JP7719482B2 - AC motor control device and control method - Google Patents

Description

Article 30, paragraph 2 of the Patent Act applies 1. Publication date of the invention (date when it became available for download from the website): May 30, 2021 2. Website address: https://www.bookpark.ne.jp/cm/ieej/detail/IEEJ-20210602X11201-PRT/ 3. Names of the people who published the invention: Masaru Hasegawa, Naoki Kawamura, Takaya Katase 4. Contents of the invention: Torque control method for a permanent magnet synchronous motor that specifies torque response

The present invention relates to a control device and control method for an AC motor.

Control devices and control methods for AC motors include the "Motor Control Device and Motor Control Method" and "AC Motor Control Device" disclosed in Patent Documents 1 and 2 below. In these types of devices, for example, current vector control is performed on the AC motor. In current vector control, the current output to each phase of the AC motor is vector-decomposed into a q-axis current component that generates torque in the AC motor and a d-axis current component that generates magnetic flux in the rotor of the AC motor, and each is controlled independently. Therefore, information on the current flowing through each phase or any phase of the AC motor is converted into d-axis current and q-axis current, then fed back and used for control (Patent Document 1: Figures 1 and 7; Patent Document 2: Figures 5 and 6). Note that throughout this specification, AC motors may be simply referred to as "motors."

Japanese Patent Application Laid-Open No. 2004-180441 JP 2014-113026 A

Yosuke Nakayama and Shinji Michiki, "Design of a Band-Rejection Filter for PMSM Vector Control in Inverter Overmodulation Region," IEEJ Transactions on Power Systems, Vol. 138, No. 11, pp. 884-893, 2018

However, AC motors controlled in this way are typically used as power sources for mobile objects, a representative example of which is an electric vehicle. Therefore, to supply AC power to an AC motor, an inverter is used, which converts DC power output from a DC voltage source such as a battery or voltage converter into AC power. The inverter is controlled by a control device, but it cannot output a voltage that exceeds the output voltage of the DC voltage source. Due to the limit value of the voltage that the inverter can output (output limit voltage), the following problems can arise with current vector control performed by the control device. Note that this discussion assumes that the AC motor is a three-phase AC motor.

(1) As shown in Figure 2 of Patent Document 1, in a voltage vector diagram showing the relationship between a stationary coordinate system representing the voltage axes of each phase (U, V, W) of a three-phase AC motor and a rotating coordinate system representing the voltage supplied to the motor in a vector space of two orthogonal axes (d and q axes), the range of voltage V_max that can be supplied to the motor is expressed as a hexagon. In other words, this hexagon represents the inverter's output voltage limit. Therefore, while it is desirable for the control device to control the inverter so that the inverter's output voltage reaches the boundaries corresponding to each side of the hexagon, this requires an algorithm that enables complex limiter control that is tailored to each side. Therefore, the time required for such calculation processing can lead to a decrease in the response speed to control commands.

(2) For this reason, in the control device of Patent Document 1, in the overmodulation region within the hexagon beyond the inscribed circle C_2 of the hexagon, the d-axis voltage shortfall Vdn beyond the inscribed circle C_2 is calculated and the q-axis current command value is corrected. The d-axis priority control and sash current control performed in such an overmodulation region tend to require complex calculations. Furthermore, in current vector control, as shown in Figures 1 and 7 of Patent Document 1, both the d-axis current and the q-axis current are generated by PI control (d-axis current controller, q-axis current controller), and this processing often involves integral calculations using an integrator or the like. Therefore, these calculation processes can ultimately lead to a decrease in the control response speed.

(3) Furthermore, as shown in the above-mentioned Non-Patent Document 1, it is known that in inverter control in the overmodulation region, various harmonics are contained in the current fed back by the control device (see, for example, Figs. 1 and 6 of Non-Patent Document 1). If these harmonics are superimposed on the feedback current as disturbances, motor control may become unstable or control performance may deteriorate. Therefore, a configuration is often adopted in which these harmonics are removed using a filter. However, such filtering generally involves integral calculations using an integrator or the like, and the delay time associated with these integral calculations can reduce the response speed of the control.

The present invention has been made to solve the above-mentioned problems, and aims to provide a control device and control method for an AC motor that can suppress a decrease in control response speed.

When the AC motor is an Interior Permanent Magnet Synchronous Motor (IPMSM), it is known that the voltage equation of the IPMSM in the dq rotating coordinate system (d-axis voltage v _d , q-axis voltage v _q ) can be expressed by the following equation (1):

R, _Ld , _Lq , and k _e are parameters specific to an AC motor, where R is the winding resistance of the coils of each phase that make up the AC motor, _Ld is the inductance of the d-axis equivalent coil, _Lq is the inductance of the q-axis equivalent coil, k _e is the back electromotive force constant, i _d is the d-axis current, i _q is the q-axis current, and ω _re is the electrical angular velocity of the output shaft of the AC motor.

If the d-axis current i _d and the q-axis current i _q are differentiated with respect to time based on the above equation (1), the following equations (2) and (3) are obtained, and the torque (output torque) τ of the output shaft of the AC motor is expressed by equation (4), where P is the number of pole pairs of the AC motor.

In this way, the mathematical model of an IPMSM can be expressed using the three equations (2) to (4) above, and its output torque τ is expressed by equation (4) above. Therefore, current vector control of an AC motor is performed based on equations (1) and (2) above. That is, in current vector control, information about the AC current (motor current) flowing through the AC motor is detected by a sensor and fed back to perform PI (proportional-integral) control so that the motor current matches the command current value, thereby obtaining the target output torque. Therefore, an IPMSM AC motor can be represented by the motor model shown in Figure 1(A).

In FIG. 1A , R is the winding resistance of the coil of each phase that constitutes the AC motor 50 (AC electric motor), and L is the inductance of the coil. s is the Laplace operator. Furthermore, v is the AC voltage applied to the coil, and i is the motor current flowing through the coil. τ is the output torque of the AC motor 50. f(i) is a predetermined transfer function that converts the AC current i flowing through each phase of the coil into the output torque τ. In the case of a dq rotating coordinate system, v is v _d,q , and i is i _d,q . In addition, in this specification and drawings, subscripts such as v _d,q and i _u,v,w may have two or three letters separated by a comma (,). This indicates that these subscripts should be added one by one to the letter to which the subscript is attached, but this notation has been omitted (for example, v _d,q means v _d and v _q ).

24 , when current vector control is performed on an AC motor 130 in a motor system 100, coordinates and phase numbers of motor currents iu _,v,w detected by a current sensor 140 are converted and then fed back to a control device 110 (i _d , i _q ). In the control device 110, a current command generator 111 generates d- and q-axis current command values i ^* _d , i ^* _q based on an externally input torque command value τ ^* , and a d-axis current PI controller 112 and a q-axis current PI controller 113 generate _d- and _q -axis voltage command values v ^* d, v ^* _q that are PI-controlled so that the feedback-input current information i d, iq approaches the d- and q-axis current command values i ^* _d , i ^* _{q .} The generated voltage command values v ^* _d , v ^* _q are converted in coordinate and phase number by a coordinate/phase number converter 115 and output to an inverter 120 as voltage command values v ^* _u,v,w . Then, AC voltages ^v _u,v,w are output from the inverter 120 that is switching controlled in accordance with the voltage command values v* _u,v,w and supplied to the AC motor 130 .

However, current vector control, which can be said to be a typical control method for such AC motors, requires PI control, and therefore requires integral calculations, which cause delays in the control process. In the configuration example shown in Figure 24, integral calculations are always performed in the d-axis current PI control unit 112 and the q-axis current PI control unit 113, and the delay time associated with these integral calculations reduces the response speed to torque commands. In other words, current vector control of AC motors has limitations in preventing a decrease in control response speed as long as it includes integral calculations using integrators, etc.

For this reason, according to the results of a computer simulation when current vector control is performed as in the comparative motor system 100 shown in Figure 24, it can be seen that, as shown in Figure 25(A), the torque (output torque) τ output from the output shaft of the AC motor 130 in response to the input torque command value τ ^* can overshoot due to a decrease in response speed (indicated by the dashed ellipse in the figure). Furthermore, as shown in Figure 25(B), it can be seen that overshoot due to response delay can also occur in the d-axis and q-axis currents i _d and i _q . Note that in the figure, the hat (^) symbol above τ indicates that τ is an estimated value obtained by calculation.

Therefore, the inventors of the present application focused on the torque differential value τdot with respect to the output torque τ and viewed the AC motor as a motor model as shown in Figure 1(B). In this figure, v (vd _,q in the dq rotating coordinate system) is the AC voltage applied to the coil of each phase that constitutes the AC motor 50 (AC motor), and g(v) is a transfer function that converts the AC voltage v applied to the coil of each phase into the torque differential value τdot. In this figure, the dot symbol "·" above τ denotes time differentiation and is synonymous with the "dot" symbol in τdot in the specification (the same applies hereinafter).

g(v) of the motor model shown in Figure 1(B) can be calculated as follows: By differentiating the above equation (4) with respect to time, the following equation (5) is obtained. Substituting the above equations (2) and (3) into this equation gives equation (6), which is then rearranged with respect to the d-axis voltage _vd and the q-axis voltage _vq to obtain equation (7) for g(v). Note that g(v) expressed by equation (7) may correspond to the "predetermined function" described in the claims.
However, A( i _d , i _q ), B( i _d , i _q ), and C( i _d , i _q ,ω _re ) are

Since the above equations (6) and (7) are linear functions with time-varying coefficients for voltage, it is possible to manage the torque differential value simply by manipulating the voltage. In other words, the output torque τ of the AC motor 50 (AC electric motor) can be controlled without current control or an integral element such as an integrator. Feedforward control of the torque differential value is also possible. Hereinafter, this type of torque response management type voltage vector control will be referred to as "this voltage vector control."

Based on such technical grounds , in order to achieve the above object, the present invention employs the technical means of claim 7. According to this means, the invention of a control method for an AC motor outputs a voltage command value of an AC voltage to be output to an AC motor to an inverter capable of supplying three-phase AC power to the AC motor, and the voltage command value is a time differential value of an output torque required of the AC motor, and is input as a command value and is generated based on the torque differential value expressed by the above equations (7) and (8), i.e., the torque differential command value.

As a result, in the invention of the AC motor control device or AC motor control method, as described above, it is possible to control the output torque of the AC motor (perform constant value control of the output torque) based on the torque differential command value without performing current control, and therefore, for example, there is no need for integral calculations in the PI control processing (reference numerals 112 and 113 in Figure 24) that are performed in general current vector control.

The present invention also employs the technical means of claim 1. According to this means, a control device for an AC motor outputs a voltage command value of an AC voltage to be output to an inverter capable of supplying three-phase AC power to the AC motor. The voltage command value is a time derivative of an output torque required of the AC motor and is input as a command value and generated based on a torque differential value expressed by the above equations (7) and (8). The control device also includes a voltage information generation unit, a coordinate conversion unit, and a voltage command value output unit, and generates rotating coordinate system voltage information corresponding to two orthogonal axes of a rotating coordinate system of the AC motor based on the torque differential value, and converts the rotating coordinate system voltage information from the rotating coordinate system to stationary coordinate system voltage information corresponding to the two orthogonal axes of a stationary coordinate system based on the rotation angle of the output shaft of the AC motor. If a voltage vector represented by the stationary coordinate system voltage information is within a range of an output voltage limit of the inverter, the stationary coordinate system voltage information is output to the inverter as a voltage command value. If the voltage vector is outside the range of the output voltage limit, the stationary coordinate system voltage information representing a changed voltage vector, which has been changed to a size within the range of the output voltage limit, is output to the inverter as a voltage command value. In some cases, the voltage command value is output via a phase number conversion unit that adapts the number of phases of the voltage command value to the number of phases of the AC motor. As a result, as constant value control of the output torque based on the torque differential command value, when the voltage vector represented by the stationary coordinate system voltage information is within the range of the inverter's output limit voltage, the stationary coordinate system voltage information is output to the inverter as a voltage command value, and when the voltage vector is outside the range of the output limit voltage, stationary coordinate system voltage information representing a changed voltage vector whose size is changed to fit within the range of the output limit voltage is output to the inverter as a voltage command value.

The present invention also employs the technical means of claim 2. According to this means, a control device for an AC motor outputs a voltage command value of an AC voltage to be output to an inverter capable of supplying three-phase AC power to the AC motor. The voltage command value is generated based on a torque differential value, which is a time differential value of output torque required of the AC motor and is input as a command value. The control device also includes a voltage information generation unit, a coordinate conversion unit, and a voltage command value output unit, and generates rotating coordinate system voltage information corresponding to two orthogonal axes of a rotating coordinate system of the AC motor based on the torque differential value, and converts the rotating coordinate system voltage information from the rotating coordinate system to stationary coordinate system voltage information corresponding to the two orthogonal axes of a stationary coordinate system based on the rotation angle of the output shaft of the AC motor. If a voltage vector represented by the stationary coordinate system voltage information is within an output voltage limit range of the inverter, the stationary coordinate system voltage information is output to the inverter as a voltage command value. If the voltage vector is outside the output voltage limit range, the stationary coordinate system voltage information representing a changed voltage vector, which has been changed to a size within the output voltage limit range, is output to the inverter as a voltage command value. In some cases, the voltage command value is output via a phase number converter that adapts the number of phases of the AC motor. When the two axes of the rotating coordinate system are a d-axis parallel to the magnetic pole direction of a rotor constituting the AC motor and a q-axis perpendicular to the d-axis, i.e., in a dq rotating coordinate system, the voltage information generator generates, as rotating coordinate system voltage information, coordinate information of a voltage vector having a magnitude up to the shortest intersection point of a line expressed by a predetermined function including "torque differential value, d-axis current, q-axis current, and angular velocity of the output shaft." This provides a voltage vector with a minimum voltage amplitude, making it possible to output a voltage command value with a minimum voltage (minimum voltage amplitude) corresponding to the torque differential command value to the inverter.

The present invention also employs the technical means of claim 3. According to this means, a control device for an AC motor outputs a voltage command value of an AC voltage to be output to an inverter capable of supplying three-phase AC power to the AC motor. The voltage command value is generated based on a torque differential value, which is a time differential value of output torque required of the AC motor and is input as a command value. The control device also includes a voltage information generation unit, a coordinate conversion unit, and a voltage command value output unit, and generates rotating coordinate system voltage information corresponding to two orthogonal axes of a rotating coordinate system of the AC motor based on the torque differential value, and converts the rotating coordinate system voltage information from the rotating coordinate system to stationary coordinate system voltage information corresponding to the two orthogonal axes of a stationary coordinate system based on the rotation angle of the output shaft of the AC motor. If a voltage vector represented by the stationary coordinate system voltage information is within an output limit voltage range of the inverter, the stationary coordinate system voltage information is output to the inverter as a voltage command value. If the voltage vector is outside the output limit voltage range, the stationary coordinate system voltage information representing a changed voltage vector, which has been changed to a size within the output limit voltage range, is output to the inverter as a voltage command value. In some cases, the voltage command value is output via a phase number converter that adapts the number of phases of the AC motor. When the two axes of the rotating coordinate system are a d-axis parallel to the magnetic pole direction of the rotor constituting the AC motor and a q-axis perpendicular to the d-axis, i.e., in a dq rotating coordinate system, the voltage information generator generates rotating coordinate system voltage information based on the torque differential, the d-axis current, the q-axis current, the angular velocity of the output shaft, and a time differential command value of the d-axis current in maximum torque per ampere (MTPA) control. This allows a voltage vector to be obtained that maximizes the output torque of the AC motor relative to the inverter's output current, making it possible to output to the inverter a voltage command value that provides the most efficient output torque corresponding to the torque differential command value.

The present invention also employs the technical means of claim 4. According to this means, a control device for an AC motor outputs a voltage command value of an AC voltage to be output to an inverter capable of supplying three-phase AC power to the AC motor. The voltage command value is generated based on a torque differential value, which is a time differential value of output torque required of the AC motor and is input as a command value. The control device also includes a voltage information generation unit, a coordinate conversion unit, and a voltage command value output unit, and generates rotating coordinate system voltage information corresponding to two orthogonal axes of a rotating coordinate system of the AC motor based on the torque differential value, and converts the rotating coordinate system voltage information from the rotating coordinate system to stationary coordinate system voltage information corresponding to the two orthogonal axes of a stationary coordinate system based on the rotation angle of the output shaft of the AC motor. If a voltage vector represented by the stationary coordinate system voltage information is within an output limit voltage range of the inverter, the stationary coordinate system voltage information is output to the inverter as a voltage command value. If the voltage vector is outside the output limit voltage range, the stationary coordinate system voltage information representing a changed voltage vector, which has been changed to a size within the output limit voltage range, is output to the inverter as a voltage command value. In some cases, the voltage command value is output via a phase number conversion unit that adapts the number of phases of the voltage vector to the number of phases of the AC motor. If the voltage vector is outside the output limit voltage range, the voltage command value output unit reduces the torque differential value until the changed voltage vector falls within the output limit voltage range. As a result, if the voltage vector is outside the output limit voltage range, that is, if an unrealizable voltage command value is requested, the torque differential value is reduced until the voltage vector falls within the output limit voltage range, allowing the voltage command value output unit to output a realizable voltage command value to the inverter without performing an integral operation.

The present invention also employs the technical means of claim 5. According to this means, a control device for an AC motor outputs a voltage command value of an AC voltage to be output to an inverter capable of supplying three-phase AC power to the AC motor. The voltage command value is generated based on a torque differential value, which is a time differential value of output torque required of the AC motor and is input as a command value. The control device also includes a voltage information generation unit, a coordinate conversion unit, and a voltage command value output unit, and generates rotating coordinate system voltage information corresponding to two orthogonal axes of a rotating coordinate system of the AC motor based on the torque differential value, and converts the rotating coordinate system voltage information from the rotating coordinate system to stationary coordinate system voltage information corresponding to the two orthogonal axes of a stationary coordinate system based on the rotation angle of the output shaft of the AC motor. If a voltage vector represented by the stationary coordinate system voltage information is within an output voltage limit range of the inverter, the stationary coordinate system voltage information is output to the inverter as a voltage command value. If the voltage vector is outside the output voltage limit range, the stationary coordinate system voltage information representing a changed voltage vector, which has been changed to a size within the output voltage limit range, is output to the inverter as a voltage command value. In some cases, the voltage command value is output via a phase number conversion unit that adapts the number of phases of the AC motor. When the two axes of the rotating coordinate system are a d-axis parallel to the magnetic pole direction of a rotor constituting the AC motor and a q-axis perpendicular to the d-axis in a dq rotating coordinate system, that is, when the voltage vector is outside the output limit voltage range, that is, when an unrealizable voltage command value is requested, the voltage command value output unit performs the following control: When, in the rotating coordinate system, a line expressed by a predetermined function including "the torque differential value, the d-axis current, the q-axis current, and the angular velocity of the output shaft" intersects with the line expressed by the predetermined function, the voltage vector extending to the intersection is determined as a modified voltage vector, and stationary coordinate system voltage information representing the modified voltage vector is output to the inverter as a voltage command value. As a result, the changed voltage vector reaches one of the six sides of the hexagon representing the range of the output limit voltage, so the voltage command value output unit can output the maximum value as a achievable voltage command value to the inverter without using, for example, limiter control that requires complex calculations, or without performing integral calculations.

The present invention also employs technical means according to claim 6. According to this means, a control device for an AC motor outputs a voltage command value of an AC voltage to be output to an inverter capable of supplying three-phase AC power to the AC motor. The voltage command value is generated based on a torque differential value, which is a time differential value of output torque required of the AC motor and is input as a command value. The control device also includes a voltage information generation unit, a coordinate conversion unit, and a voltage command value output unit, and generates rotating coordinate system voltage information corresponding to two orthogonal axes of a rotating coordinate system of the AC motor based on the torque differential value, and converts the rotating coordinate system voltage information from the rotating coordinate system to stationary coordinate system voltage information corresponding to the two orthogonal axes of a stationary coordinate system based on the rotation angle of the output shaft of the AC motor. If a voltage vector represented by the stationary coordinate system voltage information is within an output voltage limit range of the inverter, the stationary coordinate system voltage information is output to the inverter as a voltage command value. If the voltage vector is outside the output voltage limit range, the stationary coordinate system voltage information representing a changed voltage vector, which has been changed to a size within the output voltage limit range, is output to the inverter as a voltage command value. In some cases, the voltage command value is output via a phase number conversion unit that adapts the number of phases of the AC motor. When the two axes of the rotating coordinate system are a d-axis parallel to the magnetic pole direction of a rotor constituting the AC motor and a q-axis perpendicular to the d-axis in a dq rotating coordinate system, that is, when the voltage vector is outside the output limit voltage range, i.e., when an unrealizable voltage command value is requested, the voltage command value output unit performs the following control. When the most parallel side of six sides constituting a hexagon representing the range of the output limit voltage does not intersect with a line expressed by a predetermined function including "torque differential value, d-axis current, q-axis current, and angular velocity of the output shaft" in the rotating coordinate system, the voltage vector with the larger torque differential value of the two voltage vectors representing the endpoints of the most parallel side is set as a modified voltage vector, and stationary coordinate system voltage information representing the modified voltage vector is output to the inverter as a voltage command value. As a result, the changed voltage vector reaches one of the six vertices of the hexagon representing the range of the output limit voltage, so the voltage command value output unit can output the maximum value as a achievable voltage command value to the inverter without using, for example, limiter control that requires complex calculations, or without performing integral calculations.

In the seventh aspect of the present invention , constant value control of output torque based on a torque differential command value can be performed without current control, and therefore, for example, integral calculation in the PI control process (reference numerals 112 and 113 shown in FIG. 24) performed in general current vector control is not required. Therefore, since there is no delay time associated with integral calculation, it is possible to suppress a decrease in response speed to a control command.

In the invention of claim 1 , as constant value control of the output torque based on the torque differential command value, when a voltage vector represented by stationary coordinate system voltage information is within the inverter's output limit voltage range, stationary coordinate system voltage information is output to the inverter as a voltage command value, and when the voltage vector is outside the output limit voltage range, stationary coordinate system voltage information representing a changed voltage vector changed to a size that falls within the output limit voltage range is output to the inverter as a voltage command value. Therefore, not only when the voltage vector represented by the stationary coordinate system voltage information is within the output limit voltage range, but also when it is outside the output limit voltage range, the stationary coordinate system voltage information can be output to the inverter as a voltage command value within the output limit voltage range.

In the second aspect of the present invention, since the minimum value is obtained as the voltage vector, it is possible to output to the inverter a voltage command value with the minimum voltage (minimum voltage amplitude) corresponding to the torque differential command value, thereby satisfying, for example, the need to minimize the inverter output voltage or the AC motor input voltage.

In the third aspect of the present invention, a voltage vector that maximizes the output torque of the AC motor relative to the inverter output current can be obtained, making it possible to output to the inverter a voltage command value that provides the most efficient output torque corresponding to the torque differential command value. This allows for a smooth transition to MTPA control, thereby satisfying needs such as minimizing the inverter output current relative to the required torque.

In the fourth aspect of the present invention, when the voltage vector is outside the output limit voltage range, i.e., when an unrealizable voltage command value is requested, the torque differential value is reduced until the voltage vector falls within the output limit voltage range, so that the voltage command value output unit can output a realizable voltage command value to the inverter without performing an integral calculation. Therefore, since there is no delay time associated with the integral calculation, it is possible to suppress a decrease in the response speed to the control command.

In the fifth aspect of the present invention, the post-change voltage vector reaches one side of the hexagon representing the range of the output limit voltage, so that the voltage command value output unit can output the maximum value as a achievable voltage command value to the inverter without using, for example, limiter control requiring complex calculations or performing integral calculations. Therefore, since there is no delay time associated with complex calculations or integral calculations, it is possible to suppress a decrease in the response speed to control commands.

In the sixth aspect of the present invention, the post-change voltage vector reaches one vertex of a hexagon representing the range of the output limit voltage, so that the voltage command value output unit can output the maximum value as a achievable voltage command value to the inverter without using, for example, limiter control requiring complex calculations or performing integral calculations. Therefore, since there is no delay time associated with complex calculations or integral calculations, it is possible to suppress a decrease in the response speed to control commands.

These are conceptual diagrams of motor models that mathematically represent AC motors. Figure 1(A) is a model used in current vector control, and Figure 1(B) is a model that focuses on the torque differential value with respect to the output torque. Fig. 2(A) is a block diagram showing an example of the hardware configuration of a motor system to which the control device and control method for an AC motor of the present invention are applied, and Fig. 2(B) is a circuit diagram showing an example of the configuration of an inverter included in the motor system of this embodiment. FIG. 2 is a block diagram showing a first example of the configuration of control system elements that make up the motor system of the present embodiment. 4A and 4B are conceptual diagrams of the voltage vector control performed by the controller shown in FIG. 3, where FIG. 4A shows a conceptual example of a generation algorithm of a voltage command value performed in a voltage information generation unit, and FIG. 4B shows a conceptual example of a control algorithm performed in a voltage command value output unit. 5A and 5B are explanatory diagrams showing the results of a computer simulation of this voltage vector control by the controller shown in FIG. 3, where FIG. 5A shows the torque command value, output torque value, etc., and FIG. 5B shows the trajectory of the inverter output voltage. FIG. 10 is a block diagram showing a second configuration example of control system elements that make up the motor system of the present embodiment. 7 is a flowchart showing a flow of a voltage command value output process performed by the controller shown in FIG. 6 . 8A and 8B are explanatory diagrams showing the results of a computer simulation of this voltage vector control using the controller shown in FIG. 6, in which FIG. 8A shows the output torque value relative to the torque command value, FIG. 8B shows the d-axis and q-axis current values, and FIG. 8C shows the motor current value. 9A and 9B are explanatory diagrams showing the results of a computer simulation of this voltage vector control by the controller shown in FIG. 6, where FIG. 9A shows the trajectory of the inverter output voltage, and FIG. 9B shows the switching waveforms input to the switching elements of each phase that make up the inverter. 10A and 10B are explanatory diagrams showing the results of measuring the torque command value, torque estimation value, and d-axis and q-axis current values using an actual device when this voltage vector control is performed using the controller shown in FIG. 6 , where FIG. 10A shows the case of linear region operation that does not include overmodulation operation, and FIG. 10B shows the case of linear region operation that includes temporary overmodulation operation. FIG. 10 is a block diagram showing a third example of the configuration of the control system elements that make up the motor system of the present embodiment. 12 is a conceptual diagram of the voltage vector control performed by the controller shown in FIG. 11, showing a conceptual example of a generation algorithm of a voltage command value performed in a voltage information generating unit. 13A and 13B are explanatory diagrams showing the results of a computer simulation of this voltage vector control using the controller shown in FIG. 11, where FIG. 13A shows the torque command value and output torque value, FIG. 13B shows the d-axis and q-axis current values, and FIG. 13C shows the motor current value. 1 is a conceptual diagram of the voltage vector control performed by the controllers of the control system element configuration examples 1 to 3, and shows a conceptual example of a voltage command value correction algorithm performed in the voltage command value output unit. 15 is a flowchart showing the flow of a voltage command value correction process (I) corresponding to the voltage command value correction algorithm shown in FIG. 14 . 10 is a conceptual diagram of the voltage vector control performed by the controllers of the control system element configuration examples 1 to 3, and shows a conceptual example of another example of the voltage command value correction algorithm performed in the voltage command value output unit. 17 is a flowchart showing the flow of a voltage command value correction process (II) corresponding to the voltage command value correction algorithm shown in FIG. 16 . 10 is a flowchart showing the flow of integrated control processing in the voltage vector control performed by the controllers of the control system element configuration examples 1 to 3. 19A and 19B are explanatory diagrams showing the results of a computer simulation of this voltage vector control using the integrated control processing shown in FIG. 18, in which FIG. 19A shows the torque command value, output torque value, d-axis and q-axis current values, motor current values, etc., and FIG. 19B shows the three-phase voltage command values output to the inverter. 20A and 20B are explanatory diagrams showing the results of a computer simulation of this voltage vector control using integrated control processing under conditions different from those of FIG. 19, where FIG. 20A shows the torque command value, output torque value, d-axis and q-axis current values, motor current values, etc., and FIG. 20B shows the three-phase voltage command values output to the inverter. FIG. 10 is a block diagram showing a fourth example of the configuration of the control system elements that make up the motor system of the present embodiment. 22A and 22B are explanatory diagrams showing the results of a computer simulation of this voltage vector control using the controller shown in FIG. 21, where FIG. 22A shows the torque command value and output torque value, FIG. 22B shows the d-axis and q-axis current values, and FIG. 22C shows the motor current value. FIG. 23(A) is an explanatory diagram relating to an arbitrary coordinate system, and FIG. 23(B) is an explanatory diagram relating to a stationary coordinate system. FIG. 1 is a block diagram showing an example of the configuration of control system elements when current vector control is performed in a motor system. 25A and 25B are explanatory diagrams showing the results of a computer simulation when current vector control is performed, in which FIG. 25A shows the torque command value and output torque value, and FIG. 25B shows the d-axis current value and q-axis current value.

An embodiment of a motor system 10 incorporating an AC motor control device and control method of the present invention will be described below with reference to the drawings. First, the hardware that constitutes the motor system of this embodiment will be described with reference to Figures 1 to 3.

As shown in Figure 2(A), the motor system 10 is primarily composed of a controller 20, an inverter 30, a rotation angle sensor 45, an AC motor 50, etc., and is used, for example, in drive motor systems that control the traction motors of electric vehicles, electric railway vehicles, etc., and servo motor systems that control the drive motors of robots, etc.

The controller 20 is an example of a control device for an AC motor, and is typically a so-called microcomputer board (not shown) that includes, for example, a microcomputer (CPU), semiconductor memory devices such as ROM, RAM, and EEPROM, and an input/output interface. The controller 20 is connected to the inverter 30, current sensors 41, 42, and 43, a rotation angle sensor 45, and a higher-level system (not shown), and receives inputs of information on motor currents i _u,v,w supplied to the AC motor 50 and information on the rotation angle (mechanical angle) θ _rm of the shaft (output shaft) of the AC motor 50 from these sensors 41 to 43 and 45, and also receives inputs of information (command value) on the output torque τ required of the AC motor 50 from the higher-level system.

Based on this information, the controller 20 generates voltage command values v ^* _u,v,w according to the voltage vector control algorithm using control system element configuration examples 1 to 4 described below, and outputs them to the inverter 30. Therefore, the EEPROM or the like of the controller 20 pre-stores programs that implement, in software, the functions of the "voltage information generator 21, coordinate converter 22, voltage command value output unit 23, phase number converter 24, etc." (These components constitute the voltage vector control algorithm.) The inventors of the present application focus on the torque differential value τdot with respect to the output torque τ of the AC motor 50 and consider the AC motor as a motor model as shown in FIG. 1B ("dot" means that the character or symbol immediately before it is a time differential value; the same applies below). Therefore, the command value input from the higher-level system is, in principle, not the output torque (torque command value) τ ^* required of the AC motor 50, but the torque differential value (torque differential command value) τdot ^* .

The inverter 30 is a power conversion device that converts a DC voltage suitable for driving the AC motor 50 into an AC voltage corresponding to the number of phases of the AC motor 50 and outputs the converted voltage. In this embodiment, since the AC motor 50 is a three-phase AC motor, the inverter 30 converts an externally input DC voltage into an AC voltage (output voltage) v _u,v,w according to a voltage command value v ^* _u,v,w input from the controller 20 and supplies the AC voltage to the AC motor 50. To this end, the inverter 30 is connected to a DC voltage source Batt that outputs a voltage capable of driving the AC motor 50. The inverter 30 includes a switching circuit ( FIG. 2(B) ) that converts the DC voltage into an AC voltage v _u,v,w corresponding to each phase of the AC motor 50, and a control circuit (not shown) that switches the switching circuit according to the voltage command value v ^* _u,v,w . The DC voltage source Batt may be a battery itself, or a voltage converter connected to a battery when voltage step-up or step-down is required.

As shown in FIG. 2B, inverter 30 has a switching circuit in which two series-connected switching elements (upper and lower arms) are connected in parallel, the number of which corresponds to the number of phases of the AC voltage output by inverter 30. A DC voltage source Batt is connected to both ends of each upper and lower arm, and the coils of each phase of AC motor 50 are connected to the midpoint between these upper and lower arms, i.e., the connection between the switching elements. In this embodiment, since AC motor 50 is a three-phase AC motor, the switching elements are IGBTs 31-36 with FWDs, a total of six, in the upper and lower arms. The positive electrode side of DC voltage source Batt is connected to IGBTs 31, 33, and 35 of the upper arm, and the negative electrode side is connected to IGBTs 32, 34, and 36 of the lower arm. Furthermore, the terminals of the U-phase coil, V-phase coil, and W-phase coil of AC motor 50 are connected to the connection between IGBTs 31 and 32, the connection between IGBTs 33 and 34, and the connection between IGBTs 35 and 36, respectively. The switching element may be a transistor such as a power MOSFET.

The current sensors 41 to 43 are detection devices configured to be able to measure AC current, and measure the current value of AC power supplied from the inverter 30 to the AC motor 50 and output the measured value to the controller 20. For example, current information i _{u, v, w} for each of the U, V, and W phases is output from the current sensors 41 to 43 attached to the electrical wiring of each phase connecting the inverter 30 and the AC motor 50 and input to the controller 20. The current information i _{u, v, w} input from the current sensors 41 to 43 is phase-converted into d-axis and q-axis current information i _{d, q} in the controller 20 and then used.

The rotation angle sensor 45 is a detection device configured to be able to measure the rotation angle of the shaft of the AC motor 50, and is attached, for example, to the anti-load side of the shaft, and outputs angle information (mechanical angle) _θrm to the controller 20. In this embodiment, an absolute encoder or resolver that can detect the rotation angle as an absolute value is used. The absolute encoder outputs the angle information _θrm as a digital value, and the resolver outputs the angle information _θrm as an analog value (sine wave). Because the angle information _θrm input from the rotation angle sensor 45 is a mechanical angle, it is converted into an electrical angle _θre or an angular velocity _ωre in the controller 20 and used.

The AC motor 50 is, for example, an interior permanent magnet synchronous motor (IPMSM) that is powered by a supply of three-phase AC power, and has four poles P. The structure of the AC motor 50 is not specifically illustrated, but it mainly includes, for example, a cylindrical stator having multiple slots on its inner periphery around which U, V, and W phase coils are wound; a shaft (output shaft) that is connected to a load at one end (load side) and to which a rotation angle sensor 45 or the like can be connected at the other end (anti-load side); a cylindrical rotor through which the shaft passes through the center and in which multiple sets of permanent magnets are embedded circumferentially; and a case that houses these components except for both ends of the shaft (load side and anti-load side).

If the motor model of AC motor 50 is expressed in the form of a block diagram, it is represented by control system elements that are an extension of the motor model shown in Fig. 1(B). That is, as shown in Fig. 3, AC motor 50 has the functions of coordinate transformation element 51, function element 52, and integral element 53. Coordinate transformation element 51 has a function of converting a three-phase voltage vector ( _vu , _vv , _vw ) in the UVW stationary coordinate system into a voltage vector ( _vd , _vq ) in the dq rotating coordinate system. Function element 52 and integral element 53 are the same as those in the motor model of AC motor 50 described with reference to Fig. 1(B).

The motor system 10, whose hardware is configured in this manner, can be conceptualized as control system element configuration example 1, represented by the block diagram shown in Figure 3. From here on, configuration example 1 of the control system elements of the motor system 10 (10a) will be explained based on Figures 3 to 15.

[Configuration example 1 of control system elements]
As shown in FIG. 3, in a motor system 10a according to a first configuration example of control system elements (hereinafter referred to as “first configuration example”), a controller 20a has the functions of a voltage information generating unit 21, a coordinate conversion unit 22, a voltage command value output unit 23, and a phase number conversion unit 24.

As described above, the controller 20a receives a torque differential command value τdot ^* as a command value from a host system (not shown). As described with reference to FIG. 1B, when focusing on the torque differential value τdot with respect to the output torque τ, the AC motor 50 can be expressed by two control system elements: a function element of g(v) and an integral element of 1/s. The former is τdot = A( _id , _iq ) _vd + B( _id , _iq ) _vq + C( _id , _iq , _ωre ) as expressed in the above-mentioned equation (7). The latter is a pure integrator. In other words, it can be seen that g(v) expressed as τdot = A... has a linear and static relationship between the torque differential value τdot on the left side and the voltage vector ( _vd , _vq ) on the right side. It can also be understood that an integral element is included in the AC motor 50 to be controlled, whereas the controller 20a (voltage information generating unit 21, etc.) that performs the control does not include an integral element such as an integrator.

Therefore, the torque differential value τdot can be directly specified from the voltage vector ( _vd , _vq ), and vice versa. However, the inverse function v=g ^-1 (τdot) of g(v), which specifies the voltage vector ( _vd , _vq ) from the torque differential value τdot, requires two input variables _, and therefore there are countless combinations of the d _- axis voltage _vd and the q-axis voltage _vq . However, because the inverse function v=g ^-1 (τdot) is a linear function, it is unlikely to become a complex algorithm and can be realized with simpler information processing than a nonlinear function. Furthermore, as described with reference to Figures 24 and 25, current vector control always includes an integral operation (integral element), whereas the calculation process of the inverse function v=g ^-1 (τdot) does not include an integral operation (integral element). Therefore, overshoot of the output torque τ (within the dashed ellipse in Figure 25) due to the delay time associated with the integral operation in current vector control does not occur. Therefore, there is no need to perform anti-windup processing or the like to suppress the occurrence of such an overshoot in the output torque τ.

For this reason, voltage information generator 21 functions as the inverse function v ^* _d,q = g ^-1 (τdot ^* ) of equation (7) described in the section "Problems to be Solved by the Invention." In other words, when a torque differential command value τdot ^* is input from a higher-level system, and current information iu _,v,w is input from current sensors 41 to 43 and angle information _θrm is input from rotation angle sensor 45, voltage information generator 21 outputs a d-axis voltage command value v ^* _d and a q-axis voltage command value v ^* _q (rotating coordinate system voltage information) that can be calculated using the torque differential value τdot. The current information _iu,v,w and angle information _θrm are converted into current information id _,q and angular velocity _ωre in controller 20 and used.

As described above, since the voltage vector ( _vd , _vq ) is a function that inputs two variables, the selection of the d-axis voltage _vd and the q-axis voltage _vq remains arbitrary. Therefore, the voltage information generator 21 selects a voltage command value in accordance with a generation algorithm (voltage vector calculation process (I)) represented by the conceptual diagram shown in FIG. 4(A).

4A, in the voltage vector calculation process (I), for example, a torque differential line 91 obtained by the torque differential command value τdot ^* is given in the _vd - _vq plane of the dq rotating coordinate system on which the voltage vector _(vd , _vq ) (voltage vector 71) exists, and the point of intersection where the two lines intersect over the shortest distance is determined. The voltage vector 71 having a magnitude up to the intersection point is selected as the voltage command value v ^* _d,q . In other words, the voltage vector 71 whose voltage amplitude is minimum for the requested torque differential value τdot (torque differential command value τdot ^* ) is selected as the voltage command value v ^* _d,q .

More specifically, the controller 20a calculates the voltage command value v ^* _d,q according to equation (9). The operation control of the AC motor 50 performed using the voltage command value v ^* _d,q selected by the calculation of equation (9) is hereinafter referred to as "voltage amplitude minimum control." Note that A, B, and C are A(i _d , i _q ), B(i _d , i _q ), and C(i _d , i _q , ω _re ) shown in the above-mentioned equation (8).
Equation (9) can be obtained as follows:

By solving the above-mentioned equation (7) for the q-axis voltage _vq and then substituting the d-axis voltage _vd , the q-axis voltage _vq , and the torque differential value τdot with their respective command values, the following equation (10) is obtained. In the _vd - _vq plane, equation (10) is expressed by a torque differential value line 91 with a slope of -A/B. Therefore, in order for the voltage vector 71 to intersect with this torque differential value line 91 at the shortest distance, the voltage vector 71 must intersect with this line 91 at a right angle (be perpendicular). In other words, the voltage vector 71 only needs to satisfy equation (11). Therefore, from the above-mentioned equation (9) obtained by solving the simultaneous equations of these two equations (10) and (11), the d-axis voltage command value v ^* _d and the q-axis voltage command value v ^* _q (dq-axis voltage command values v ^* _d,q ) can be obtained as the voltage vector 71 with the smallest voltage amplitude.

The coordinate conversion unit 22 has a function of converting a voltage vector (v _d , v _q ) in the dq rotating coordinate system into a voltage vector (v _α , v _β ) in the αβ stationary coordinate system. Here, the coordinate conversion unit 22 converts the d-axis voltage command value v ^* _d and the q-axis voltage command value v ^* _q output from the voltage information generation unit 21 into an α-axis voltage command value v ^* _α and a β-axis voltage command value v ^* _β (stationary coordinate system voltage information) and outputs them. The conversion from the rotating coordinates to the stationary coordinates uses conversion formulas (v ^* _α = v ^* _d cos θ _re - v ^* _q sin θ _re , v ^* _β = v ^* _d sin θ _re + v ^* _q cos θ _re ) derived from a known conversion matrix. Therefore, in addition to the d-axis voltage command value v ^* _d and the q-axis voltage command value v ^* _q input from the voltage information generation unit 21, the coordinate conversion unit 22 also converts angle information _θrm input from the rotation angle sensor 45 into an electrical angle _θre and uses the converted information.

The voltage command value output unit 23 has a function (limiter function) of changing the voltage vector (v ^* _α , v ^* _β ) consisting of the α-axis voltage command value v ^* _α and the β-axis voltage command value v ^* _β input from the coordinate conversion unit 22 to a magnitude that falls within an output limit range 23a, which is the limit value of the voltage that the inverter 30 can output, and outputting the α-axis voltage command value v ^** _α and the β-axis voltage command value v ^** _β (voltage command values). The inverter 30 cannot output a voltage that exceeds the output voltage of the DC voltage source Batt input to the inverter 30. In other words, the output limit range 23a of the inverter 30 is predetermined based on its specifications and the output voltage of the DC voltage source Batt input. Therefore, it is possible to determine the output limit range 23a having a regular hexagonal shape as shown in the conceptual diagram of FIG. 4(B).

The regular hexagon (hereinafter referred to as the "hexagon") representing this output limit range 23a is obtained by extending three axes (U-axis, V-axis, and W-axis) that divide the coordinate plane of the αβ stationary coordinate system into thirds at 120-degree intervals circumferentially around the intersection of the orthogonal α-axis and β-axis toward the intersection, creating six lines extending in a "*" shape. These six lines are then connected at the same length (maximum output voltage). The range within the hexagon (light gray area) that extends beyond the inscribed circle of this hexagon (dashed circle shown in Figure 4(B)) is the so-called overmodulation region. Current vector control configured as shown in Figure 24 enables voltage vector control that reaches this overmodulation region by performing complex calculations such as d-axis priority control and sash current control.

For more information on d-axis priority control, see, for example, Takahashi Kenji, Oishi Kiyoshi, and Ueno Toshiyuki, "A Driving Method for Interior Permanent Magnet Synchronous Motors with Priority on d-Axis Voltage," IEEJ Transactions on Motors, Vol. 131, No. 9, pp. 1103-1111, 2011. For more information on cross-coupling control, see, for example, Makishima Shingo, Uezono Keiichi, and Nagai Masao, "Verification and Consideration of Motor Control Response in Voltage Saturation States," IEEJ Transactions on Motors, Vol. 130, No. 5, pp. 663-670, 2010.

The voltage command value output unit 23 determines whether the voltage vector (v ^* _α , v ^* _β ) in the αβ stationary coordinate system input from the coordinate conversion unit 22 is outside the output limit range 23a of the inverter 30, and if it determines that it is not outside the output limit range 23a (within the range), it outputs the input voltage vector (v ^* _α , v ^* _β ) as is to the phase conversion unit 24 as the α-axis voltage command value v ^** _α and the β-axis voltage command value v ^** _β , and if it determines that it is outside the output limit range 23a (not within the range), it changes the voltage vector (v ^* _α , v ^* _β ) to a size that falls within the output limit range 23a, and then outputs the changed voltage vector to the phase conversion unit 24 as the α-axis voltage command value v ^** _α and the β-axis voltage command value v ^** _β . In addition, for the convenience of distinguishing between the command value input to the voltage command value output unit 23 and the command value output from the voltage command value output unit 23, the latter command value is marked with " ^** ".

Since output limit range 23a is predetermined based on the specifications of inverter 30 and the output voltage of DC voltage source Batt, each side of the hexagon of output limit range 23a can be specified as a linear function on the coordinate plane of the αβ stationary coordinate system. By solving the equations representing the lines represented by voltage vector 71 as simultaneous equations for each equation representing each side of the hexagon, it can be seen that there is one intersection for each solution. If there are two or more solutions, there will also be two or more intersections; if there is no solution, there will be no intersection.

For this reason, first, a process is performed to calculate the intersection point where the six linear functions representing each side of the hexagon of output limit range 23a intersect with the line segment represented by voltage vector 71, and then a process is performed to determine whether such an intersection point has been found (whether an intersection point exists). If an intersection point exists, it is determined that voltage vector 71 is outside (not within) the hexagon of output limit range 23a; if no intersection point exists, it is determined that voltage vector 71 is within the hexagon of output limit range 23a.

The "range of output limit range 23a" does not include the sides and vertices of the hexagon that represents the range of output limit range 23a. Therefore, if the coordinates representing the tip position of voltage vector 71 overlap with a side or vertex of the hexagon, that is, if an intersection exists, it is determined to be outside the range of output limit range 23a. Note that the process of "changing the voltage vector (v ^* _α , v ^* _β ) to a size that fits within the range of output limit range 23a," which is performed when it is determined to be outside the range of output limit range 23a, will be described later in configuration example 2 of the control system element.

The phase number converter 24 has a function of converting a two-phase voltage vector (v _α , v _β ) in the αβ stationary coordinate system into a three-phase voltage vector (v _u , v _v , v _w ) in the UVW stationary coordinate system. Here, the phase number converter 24 converts the α-axis voltage command value v ^** _α and the β-axis voltage command value v ^** _β output from the voltage command value output unit 23 into a U-phase voltage command value v ^* _u , a V-phase voltage command value v ^* _v , and a W-phase voltage command value v ^* _w , and outputs them (voltage command values v ^* _u,v,w ). The UVW-phase voltage command values v ^* _u,v,w are output from the controller 20a and input to the inverter 30. As a result, the inverter 30 generates (converts) AC voltages v u,v,w by switching the IGBTs 31 to 36 in accordance with the voltage command values v ^* _u,v,w input from the controller 20, as described above, and supplies the AC voltages v _u,v,w to the AC motor 50.

As described above, in the motor system 10a of configuration example 1, the controller 20a includes a voltage information generating unit 21, a coordinate conversion unit 22, a voltage command value output unit 23, and a phase number conversion unit 24. The voltage information generating unit 21 generates dq-axis voltage command values v ^* d,q (rotating coordinate system voltage information) in the dq rotating coordinate system of the AC motor 50 from the inverse function v ^* _d,q = g ^-1 (τdot ^* ) of equation (7) based on the torque differential command value τdot* _input from an external device such as a higher-level system. The coordinate conversion unit 22 converts the dq-axis voltage command values v ^* _{d,q (} rotating coordinate system voltage information) into αβ-axis voltage command values v ^* _α,β (stationary coordinate system voltage information) in the αβ stationary coordinate system based on the electrical ^angle θre of the output shaft of the AC motor 50. Furthermore, when the voltage vector (v ^* _α , v ^* _β ) represented by the αβ-axis voltage command value v ^* _α,β is within the output limit range 23a of the inverter 30, the voltage command value output unit 23 outputs the αβ-axis voltage command value v ^* _α,β to the inverter 30 via the phase number conversion unit 24 as the αβ-axis voltage command value v ^** _α,β .

As a result, in the voltage amplitude minimum control by the controller 20a, constant value control of the output torque τ based on the torque differential command value τdot ^* is possible without performing current control, and therefore, for example, integral calculation in the PI control process (reference numerals 112 and 113 shown in FIG. 24) performed in general current vector control is not required. Therefore, since there is no delay time associated with integral calculation, it is possible to suppress a decrease in the response speed to a control command.

Such technical effects can be seen, for example, from the results of computer simulation of this voltage vector control by the controller 20a, which show the response characteristics of the output torque and the like shown in FIG. 5(A) and the trajectory of the output voltage (AC voltage v _u,v,w ) for the inverter 30 shown in FIG. 5(B), which will be described with reference to FIG. 5.

In the simulation, as shown in the dashed rectangular box at the bottom of FIG. 3 , a subtraction unit 25 and a torque control unit 26 are provided before the voltage information generation unit 21, and a torque differential command value τdot ^* is input to the voltage information generation unit 21. The torque command value τ ^* output from a higher-level system (not shown) and a calculated torque estimate value τhat are input to the subtraction unit 25, and the result of subtracting the torque estimate value τhat from the torque command value τ ^* (the estimated deviation of the output torque τ from the torque command value τ ^* ) is input to a torque control unit 26 of a control gain K. In other words, the torque differential command value τdot ^* input to the voltage information generation unit 21 is expressed as τdot ^* = K(τ ^* - τhat). The torque control unit 26 can control the magnitude of the torque differential command value τdot ^* and the rise time of the output torque τ by changing the magnitude of the control gain K.

In the figure, the hat symbol "^" above τ denotes an estimated value, and is expressed as "hat" in the specification. Note that τhat is calculated from the rotation speed of the AC motor 50 and the motor currents iu _,v,w , etc. In this simulation, an arbitrary value is input as a setting condition for the gain K of the torque control unit.

The conditions for this simulation are as follows:
Rotation speed of AC motor 50: 1800 min ⁻¹ (constant)
Torque control unit control gain K: 2000 rad/sec (time constant 500 μsec)
Torque command value τ ^* : Increases from 0 N·m to 1 N·m

In this simulation, the torque command value τ ^* was set to a value (1 N·m) that could be achieved within the output limit range 23a of the inverter 30. Therefore, as shown in FIG. 5A, it was confirmed that the output torque τ rose within the hexagon (output limit range 23a) within the first calculation processing time (500 μsec) of the controller 20a. It was also confirmed that no overshoot occurred in the output torque τ. The rise time of the output torque τ is the time required from the start of control after input of the torque command value τ ^* until it reached 63.2% of the target torque. The time constant in the above condition is the design value of the rise time.

Furthermore, the output voltages v _α and v _β of the inverter 30 shown in FIG. 5B are αβ-axis voltages v _{α and β} obtained by transforming the output voltages v _u,v,w of the inverter 30 into an αβ stationary coordinate system. The trajectories of the output voltages v _α and v _β of the inverter 30 rotate counterclockwise in a circular motion (reference numeral 71a) before the input of the torque command value τ ^* , rise upon the input of the torque command value τ ^* (reference numeral 71b), and then return to a circular motion (reference numeral 71c) after settling. Even at the rising peak 71b', the output torque τ remains within the hexagon (output limit range 23a), confirming that the theoretical response performance is achieved. Note that, because the AC motor 50 is a synchronous motor, the trajectories of the output voltages v _α and v _β rotate in synchronization with the rotational speed of the AC motor 50 (the frequency of the three-phase AC voltages v _u,v,w ).

5A and 5B, it can be seen that the output torque τ does not overshoot the torque command value τ ^* . That is, in this voltage vector control, the voltage information generating unit 21 (the calculation process of the inverse function v = g ^-1 (τdot)) and the coordinate conversion unit 22 do not include integral calculations (integral elements). Therefore, even if anti-windup processing or the like is not provided, no overshoot that may be caused by delay times associated with calculation processing occurs in the output torque τ.

Incidentally, when the voltage vector (v ^* _α , v ^* _β ) is outside the output limit range 23a, the voltage command value output unit 23 needs to change the magnitude to fall within the output limit range 23a. This is because the inverter 30 cannot output a voltage that exceeds the output limit range 23a to the AC motor 50. Therefore, such a case is addressed by a second configuration example of the control system elements of the motor system 10 (10b) shown in FIG. 6. Next, a second configuration example of the control system elements of the motor system 10 (10b) will be described with reference to FIGS. 6 to 10.

[Configuration example 2 of control system elements]
As shown in Fig. 6, in motor system 10b of configuration example 2 of control system elements (hereinafter referred to as "configuration example 2"), controller 20b has the functions of voltage information generator 21, coordinate converter 22, voltage command value output unit 23, phase number converter 24, subtractor 25, and torque controller 26. In other words, controller 20b differs from controller 20a of motor system 10a of configuration example 1 (Fig. 3) in that the functions of subtractor 25 and torque controller 26 have been added to controller 20b. Therefore, in the description of controller 20b, the same components as those in controller 20a will be denoted by the same reference numerals and description thereof will be omitted.

The subtraction unit 25 and torque control unit 26 are almost the same as those used in the simulation of the motor system 10a in configuration example 1, except that the control gain K of the torque control unit 26 can be controlled externally, which will now be explained.

The torque control unit 26 is a torque control unit that can externally control the control gain K, and in the controller 20b of configuration example 2, a control input that can increase or decrease the gain K is connected to the voltage command value output unit 23, so that the gain K can be controlled by the voltage command value output unit 23. As a result, information on the estimated deviation (τ ^* - τhat) of the output torque τ from the torque command value τ ^* output from the subtraction unit 25 is input to the torque control unit 26, and the deviation is multiplied by the gain K and output as the torque differential command value τdot ^* to the voltage information generation unit 21 (τdot ^* = K(τ ^* - τhat)). Since the magnitude of the gain K is controlled by the voltage command value output unit 23, the voltage command value output unit 23 can decrease or increase the torque differential command value τdot ^* .

In this way, in controller 20b, torque command value τ ^* is input from an external higher-level system, and the estimated deviation of output torque τ from torque command value τ ^* is reduced or increased by gain K, and the result is input as torque differential command value τdot ^* to voltage information generating unit 21. Therefore, by controlling this gain K, voltage command value output unit 23 can change the voltage vector (v ^* _α , v ^* _β ) converted by coordinate conversion unit 22 to a magnitude that falls within output limit range 23a when the voltage vector is outside output limit range 23a.

As described in configuration example 1, the voltage command value output unit 23 determines whether the voltage vector (v ^* _α , v ^* _β ) in the αβ stationary coordinate system input from the coordinate conversion unit 22 is within the output limit range 23a of the inverter 30, and if it determines that the voltage vector is outside the output limit range 23a, changes the voltage vector (v ^* _α , v ^* _β ) to a size that falls within the output limit range 23a. Then, it outputs to the phase number conversion unit 24 an α-axis voltage command value v ^** _α and a β-axis voltage command value v ^** _β that represent the changed voltage vector.

Specifically, for example, controller 20b performs a voltage command value output process as shown in FIG. 7. This process is stored in advance in the EEPROM of controller 20b as a program that implements the functions of voltage command value output unit 23 in software. Similarly, programs that implement the functions of the subtraction unit 25 and torque control unit 26 in software are also stored in advance in the EEPROM of controller 20b.

As shown in FIG. 7, in the voltage command value output process, after voltage vector calculation processing is performed in step S101, it is determined in step S103 whether or not the voltage is outside the output limit range 23a of the inverter 30. This is the same as what was described in configuration example 1. In other words, the voltage vector calculation processing in step S101 is the process of calculating the intersection where the six linear functions representing each side of the hexagon of the output limit range 23a intersect with the line segment represented by the voltage vector 71, and the determination processing in step S103 is the process of determining whether or not such an intersection has been found (whether or not an intersection exists), i.e., whether or not the voltage vector 71 is outside the hexagon of the output limit range 23a (is not within the range).

If this determination process determines that the voltage vector 71 is outside the hexagonal range of the output limit range 23a (S103; Yes), the subsequent gain control information output process (S105) outputs control information to the torque control unit 26 to lower the gain K. The process then returns to step S101, where the voltage vector calculation process (S101) is performed, followed by the determination process (S103). When the torque control unit 26 lowers the gain K, the voltage vector 71 becomes smaller than it was in the previous calculation. Therefore, by repeating the gain control information output process (S105) → voltage vector calculation process (S101) → determination process (S103), the "voltage vector 71 with the minimum voltage amplitude" calculated using the aforementioned equation (9) is changed by the voltage command value output unit 23 to a size that fits within the output limit range 23a.

Then, when it is determined that the voltage vector 71 is within the hexagonal range of the output limit range 23a (S103; No), an αβ-axis voltage command value output process (S107) outputs an α-axis voltage command value v ^** _α and a β-axis voltage command value v ^{** β} representing the changed voltage vector (changed voltage vector) ( _v * _α , v ^* _β ) to the phase number conversion unit 24. Note that when the voltage vector is within the output limit range 23a from the beginning, the gain control information output process (S105) is not performed, and therefore the input voltage vector (v ^* _α , v ^* _β ) is output as is to the phase number conversion unit 24 as the α-axis voltage command value v ^** _α and the β-axis voltage command value v ^** _β .

As described above, in the motor system 10b of configuration example 2, when the voltage vector 71 is outside the output limit range 23a, the voltage command value output unit 23 decreases the torque differential command value τdot ^* by lowering (reducing) the gain K of the torque control unit 26 until the changed voltage vector falls within the output limit range 23a. As a result, when the voltage vector 71 is outside the output limit range 23a, that is, when an unrealizable αβ-axis voltage command value v ^* _α,β is requested, the torque differential command value τdot ^* is decreased until it falls within the output limit range 23a. The process of repeating the gain control information output process (S105), voltage vector calculation process (S101), and determination process (S103) is performed by the subtraction unit 25 and the torque control unit 26, and therefore the only control parameter is the gain K. Therefore, focusing on the control parameters, the output torque τ is expressed as τ = τ ^* (1 - e ^(-Kt) ), where t represents time.

Therefore, it can be seen that the output torque τ does not overshoot the torque command value τ ^* and is easy to manage. Furthermore, since no integral element is included, for example, processing based on past history is not performed. Therefore, in the motor system 10b of configuration example 2, it is possible to output an αβ-axis voltage command value v ^** _α,β that can be immediately realized to the inverter 30 via the phase number conversion unit 24. Furthermore, since there is no delay time associated with integral calculation, it is possible to suppress a decrease in the response speed to a control command.

Such technical effects can be seen, for example, from the results of computer simulation of this voltage vector control by the controller 20b, which show the response characteristics of the output torque and the like shown in FIG. 8 and the trajectory of the output voltage (AC voltages v _u,v,w ) for the inverter 30 shown in FIG. 9, which will be described with reference to FIGS. 8 and 9.

The conditions for this simulation are as follows:
Rotation speed of AC motor 50: 3600 min ⁻¹ (constant)
Torque control unit control gain K: 2000 rad/sec (time constant 500 μsec)
Torque command value τ ^* : Increases from 0 N·m to 3 N·m

In Figures 8 and 9, control by controller 20b of Configuration Example 2 and control by controller 20a of Configuration Example 1 are used for different periods. That is, the simulation results shown in Figures 8 and 9 were obtained by controlling the first 3 msec period from the start of control using controller 20b (Configuration Example 2) and the second half of the period thereafter using controller 20a (Configuration Example 1). Therefore, these simulation results confirm the response characteristics of controller 20b of Configuration Example 2 during the first 3 msec period from the start of control, and that at the boundary between the first and second halves, there is a seamless transition from Configuration Example 2 to Configuration Example 1 without generating noise or vibration. Note that because the functions of controllers 20a and 20b can be realized using the same hardware, such control switching can be handled via software.

In this simulation, the torque command value τ ^* was set to a value (3 N·m) that can be achieved by temporarily using the overmodulation region (the light gray region shown in FIG. 4B) within the inverter 30's output limit range 23a. Therefore, even under these conditions, where overshooting of the output torque τ is likely to occur under conventional current vector control, the waveforms of the output torque τ and the estimated torque value τhat were confirmed to be free of overshoot under this voltage vector control, as shown in FIG. 8A. Furthermore, even when switching control from Configuration Example 2 to Configuration Example 1, the waveforms of the output torque τ and the estimated torque value τhat were confirmed to be free of noise. Note that FIG. 8B shows the waveforms of the d- and q-axis currents i _{d and q} , and FIG. 8C shows the waveform of the motor current |i|. |i| is calculated by taking the square root (the square root of the sum of squares) of the U-, V-, and W-phase motor currents i _{u, v, and w} . It can be seen that there is no overshoot or noise due to switching in the waveforms of the dq-axis current i _d,q and the motor current |i|.

9A, the trajectory of the output voltage of the inverter 30 is the same as that described in Configuration Example 1, and the output voltages _vα and _vβ of the inverter 30 shown in the figure are obtained by transforming the output voltages vu _,v,w of the inverter 30 into the αβ stationary coordinate system. In this simulation, the torque command value τ ^* is intentionally set to a value (3 N·m) that temporarily uses the overmodulation region within the output limit range 23a of the inverter 30. Therefore, the portion (reference numeral 71d) where the output voltages _vα and _vβ of the inverter 30 reach the overmodulation region traces a trajectory that follows the edge of the hexagon of the output limit range 23a, but this does not indicate that an overshoot occurs in the output torque τ.

9B, when the waveforms of the switching signals (switching waveforms) input to the IGBTs 31 to 36 of the inverter 30 are examined, it can be seen that there is a period during which no switching signals are input to the IGBTs 31 to 36. This period is the time during which the output voltages v _α and v _β of the inverter 30 reach the overmodulation region. In the figure, on the vertical axis of each phase, "1.0" represents the on-timing of the IGBTs 31, 33, and 35 of the upper arm, and "0.0" represents the on-timing of the IGBTs 32, 34, and 36 of the lower arm.

Experiments were also conducted on the motor system 10b of configuration example 2 using an actual machine, and the results will be described with reference to Figure 10. Figure 10(A) shows the experimental results for linear region operation without overmodulation operation, and Figure 10(B) shows the experimental results for linear region operation with temporary overmodulation operation. These experimental results show the waveforms of the torque command value τ ^* , torque estimate value τhat, d-axis current _id , and q-axis current _iq . In this experiment, the responsiveness of the output torque was confirmed from the torque estimate value τhat instead of the output torque τ.

The conditions for this experiment are as follows:
Rotation speed of AC motor 50: 1000 min ⁻¹ (constant)
Control gain K of torque control section: 2000 rad/sec (time constant 500 μsec) in FIG. 10(A), 5000 rad/sec (time constant 200 μsec) in FIG. 10(B)
Torque command value τ ^* : Increases from 0 N·m to 2 N·m

As shown in Figure 10(A), in the case of linear region operation that does not include overmodulation operation, it was confirmed from the waveform of the estimated torque value τhat that the output torque responds to the torque command value τ ^* with almost no overshoot. In addition, it was confirmed that the time constant obtained from the waveform of the estimated torque value τhat is 600 μsec, which is almost close to the designed value of the rise time (time constant 500 μsec). Therefore, it was confirmed that torque response can be managed in the linear region that does not include overmodulation operation within the output limit range 23a.

As shown in Figure 10(B), in the case of linear operation including temporary overmodulation operation, the waveform of the estimated torque value τhat shows that the output torque increases linearly toward the torque command value τ ^* , but responds without overshoot. Therefore, within the output limit range 23a, it was confirmed that the response is without overshoot even when the system temporarily transitions to overmodulation operation. It can also be seen that the waveform of the estimated torque value τhat begins to rise approximately 200 μsec after the start of control. Therefore, although the settling time of the estimated torque value τhat is 800 μsec in this figure, excluding this approximately 200 μsec delay, it can be seen that the output torque rises at a value close to the design value, similar to the waveform of the estimated torque value τhat shown in Figure 10(A).

[Configuration example 3 of control system elements]
Next, a third configuration example (hereinafter referred to as "third configuration example") of the control system elements of the motor system 10 (10c) will be described with reference to Figures 11 to 20. In the motor system 10a of the first configuration example, the voltage information generator 21 selected the d-axis and q-axis voltage command values v ^* _d,q in accordance with the voltage vector calculation process (I) (the generation algorithm represented by the conceptual diagram shown in Figure 4(A)). As described above, the voltage vector ( _vd , vq) of the d-axis and q-axis voltage command values v ^* _d and v ^* _q obtained by the voltage information generator ₂₁ is a function that inputs two variables, and therefore the selection of the d-axis voltage _vd and the q-axis voltage _vq remains arbitrary.

Incidentally, in a synchronous motor such as the AC motor 50, copper loss in the coils of each phase is minimized when the AC motor 50 is driven at a phase angle of the motor currents iu _,v,w that maximizes the output torque τ, and therefore, MTPA (Maximum Torque Per Ampere) control, which maximizes the output torque τ with respect to the motor currents iu _,v,w, is often used in drive motor systems, servo motor systems, etc. For example, when a higher-level system controls the operation of the AC motor 50 using MTPA control, the controller 20 acquires the d-axis current command value i ^* _d as control information from the higher-level system, and thereby it becomes possible to select the dq-axis voltage command values v ^* _d,q in the same manner as the generation algorithm represented by the conceptual diagram shown in FIG. Therefore, as another example of the selection of the dq-axis voltage command values v ^* _d,q by the voltage information generating unit 21 in the case of voltage amplitude minimum control by the controller 20a of configuration example 1, we will explain the motor system 10c of configuration example 3, which can be applied when the operation control of the AC motor 50 is performed using MTPA control.

As shown in FIG. 11, in motor system 10c of configuration example 3, controller 20c has the functions of voltage information generator 21, coordinate converter 22, voltage command value output unit 23, phase number converter 24, subtractor 25, torque controller 26, subtractor 27, current controller 28, and voltage information generator 29. In other words, controller 20c differs from motor system 10a of configuration example 1 (FIG. 3) in that it has the additional functions of subtractor 25, torque controller 26, subtractor 27, current controller 28, and voltage information generator 29. Furthermore, motor system 10b of configuration example 2 (FIG. 6) differs from controller 20b in that it has the additional functions of subtractor 27, current controller 28, and voltage information generator 29. Therefore, in the description of controller 20c, the same components as controllers 20a and 20b will be designated by the same reference numerals and will not be described again.

The subtraction unit 27 is the same as that used in the simulation of the motor system 10a of Configuration Example 1, but the information input thereto is different. The d-axis current command value i ^* _{d_MTPA} and the d-axis current _id in the MTPA control output from a higher-level system (not shown) are input to the subtraction unit 27, and the result of subtracting the d-axis current _id from the d-axis current command value i ^* _{d_MTPA} (the deviation of the d-axis current _id from the d-axis current command value i ^* _{d_MTPA} ) is input to the current control unit 28.

The current control unit 28 is a control unit having a control gain G, and outputs the deviation of the d-axis current i_d input from the subtraction unit ₂₇ multiplied by the gain G as a d _- axis current differential command value i_d ( ^{i_d* d} ₌ G( ^i_d_MTPA - _{i_d} )). The magnitude of the gain G is set in advance. The d-axis current differential command _value ^i_d output from _the current control unit 28 is input to the voltage information generation unit 29 and used to calculate _the d-axis voltage command value ^v_d .

The voltage information generating unit 29 functions as an inverse function v ^* _d = h ^-1 (idot ^* _d ) of equation (15) described later. In other words, when a d-axis current command value i ^* _{d_MTPA} is input from a higher-level system performing MTPA control, a d-axis voltage command value v ^* _d that can be calculated using the d-axis current differential command value i ^* _d is output from the voltage information generating unit 29. The d-axis voltage command value v ^* _d output from the voltage information generating unit 29 is input to the voltage information generating unit 21 and the coordinate conversion unit 22. Note that the d-axis current differential command value i ^* _d is obtained by differentiating the d-axis current command value i ^* _{d_MTPA} with respect to time, and is therefore a time differential command value of the d-axis current in MTPA control.

When the voltage information generating unit 21 receives the d-axis voltage command value v ^* _d from the voltage information generating unit 29, the voltage information generating unit 21 selects a voltage command value in accordance with the generation algorithm (voltage vector calculation process (II)) represented by the conceptual diagram shown in FIG. 12.

12 , in the voltage vector calculation process (II), for example, a torque differential line 92 obtained from the torque differential command value τdot ^* is given in the _vd - _vq plane of the dq rotating coordinate system on which the voltage vector ( _vd , _vq ) (voltage vector 72) exists. In the voltage vector calculation process (I) in the above-described configuration example 1, the point of intersection where the two intersect with each other at the shortest distance is found, and the voltage vector 71 having a magnitude up to the point of intersection is selected as the voltage command value v ^* _d,q .

However, in this case, the d-axis voltage command value v ^* _d output from the voltage information generator 29 is input to the voltage information generator 21, and therefore one of the two variables, the d-axis voltage vd, of the voltage vector ( _vd , _vq ) of the d-axis voltage command value v ^* _d and the q-axis voltage command value v ^* _q obtained by the voltage information generator ₂₁ is determined. Therefore, the other q-axis voltage _vq is also determined, and the voltage vector 72 suitable for MTPA control is selected as the voltage command value v ^* _d,q .

More specifically, the controller 20c calculates the voltage command value v ^* _d,q according to equation (12). The operation control of the AC motor 50 performed using the voltage command value v ^* _d,q selected by the calculation of equation (12) is hereinafter referred to as "current amplitude minimum control." Note that A, B, and C are A(i _d , i _q ), B(i _d , i _q ), and C(i _d , i _q , ω _re ) shown in the above-mentioned equation (8).
Equation (12) can be obtained as follows.

When some of the variables in the above-mentioned equations (7) and (2) are replaced with command values, the following equations (13) and (14) are obtained. The d-axis current differential command value i ^* _d output from the current control unit 28 is expressed by equation (15), as explained above, and therefore, by solving equations (13) and (14) for the d-axis voltage command value v ^* _d and the q-axis voltage command value v ^* _q , the above-mentioned equation (12) is obtained. Note that in the following equation (15), the d-axis current command value i ^* _{d_MTPA} is expressed as "i ^* _d ".

As described above, in the motor system 10c of configuration example 3, in the dq rotating coordinate system, the voltage information generator 21 generates the d-axis voltage command value v* _d and the q-axis voltage command value v ^* _q based on the torque differential value τdot, the d-axis current id, the q-axis current _iq , the angular velocity _ωre , and _the d-axis current differential command value idot ^* _d ^. This makes it possible to obtain a voltage vector ( _vd , _vq ) that maximizes the output torque τ of the AC motor 50 for the output current _iu,v,w of the inverter 30, and therefore makes it possible to output the voltage command values αβ-axis voltage command values v ^** _α,β that maximize the efficiency of the output torque τ corresponding to the torque differential command value τdot ^* to the inverter 30 via the phase number converter 24. Therefore, the motor system 10c of configuration example 3 can satisfy needs such as minimizing the output current iu _,v,w of the inverter 30 for the torque command value τ ^* (required torque). Furthermore, since the voltage vector 72 suitable for the MTPA control is selected as the voltage command value v ^* _d,q , a smooth transition to the MTPA control is possible.

This technical effect can be seen, for example, from the response characteristics of the output torque, etc., obtained as a result of a computer simulation of this voltage vector control by controller 20c, as shown in Figure 13, which will be explained with reference to Figure 13.

The conditions for this simulation are as follows:
Rotation speed of AC motor 50: 1000 min ⁻¹ (constant)
Torque control unit control gain K: 5000 rad/sec (time constant 1 msec)
Control gain G of current control section: 1000 rad/sec (time constant 1 msec)
Torque command value τ ^* : Increases from 0 N·m to 4 N·m

13 , control by the controller 20c of Configuration Example 3 and control by the controller 20a of Configuration Example 1 are used for different periods. That is, the simulation results shown in FIG. 8 were obtained by controlling the first 1.2 msec period from the start of control using the controller 20a (Configuration Example 1) and controlling the second 1.2 msec period thereafter using the controller 20c (Configuration Example 3). Therefore, the simulation results confirm the response characteristics of the controller 20a of Configuration Example 1 during the first 1.2 msec period from the start of control. At the boundary between the first and second halves and thereafter, a seamless transition from Configuration Example 1 to Configuration Example 3 occurs without fluctuations in the output torque τ, noise, vibration, or the like. The simulation results also confirm the response characteristics of the d-axis current i _d relative to the d-axis current command value i ^* _d under MTPA control. Note that the functions of the controllers 20a and 20c can be realized by the same hardware, so such control switching can be handled by software.

In this simulation, the torque command value τ ^* was set to a value (4 N·m) that can be achieved by temporarily using the overmodulation region within the output limit range 23a of the inverter 30. Therefore, the response characteristics of the controller 20a of Configuration Example 1 during the 1.2 msec period from the start of control confirmed that the output torque τ increased linearly toward the torque command value τ ^* without overshooting. Furthermore, even when switching control from Configuration Example 1 to Configuration Example 3, it was confirmed that no noise or other artifacts were present in the waveforms of the output torque τ or the estimated torque value τhat. Furthermore, immediately after the control switchover, the d-axis current command value i ^* _d dropped sharply in a stepwise manner, and in response, the d-axis current _id approached the d-axis current command value i ^* _d , confirming a smooth transition to MTPA control.

Here, an algorithm of the voltage command value output unit 23 that can be commonly applied to the motor systems 10a to 10c of the first to third configuration examples will be described with reference to FIGS.
[Modification 1 of Configuration Examples 1 to 3]

As described above, the voltage command value output unit 23 has a limiter function of changing the voltage vector (v ^* _α , v ^* _β ) consisting of the α-axis voltage command value v ^* _α and the β-axis voltage command value v ^* _β input from the coordinate conversion unit 22 to a magnitude that falls within the output limit range 23a, which is the limit value of the voltage that can be output by the inverter 30, and outputting the α-axis voltage command value v ^** _α and the β-axis voltage command value v ^** _β . Therefore, when it is determined that the voltage vector (v ^* _α , v ^* _β ) is outside the output limit range 23a, the controller 20b of configuration example 2 uses an algorithm (FIG. 7; voltage command value output process) in which the voltage command value output unit 23 reduces the gain K of the torque control unit 26 until the changed voltage vector falls within the output limit range 23a.

However, as shown in FIG. 14, even if it is determined that the voltage vectors 71 and 72 obtained using the above-described equations (9) and (12) are outside the output limit range 23a, they may be able to fall within the output limit range 23a by changing them to another operating point on the torque differential line 93. Therefore, in the voltage command value output process by the voltage command value output unit 23 of configuration example 2, the voltage vectors are simply changed to a size that falls within the output limit range 23a, but this algorithm is not necessarily appropriate. For example, if there is a point where the torque differential line 93 intersects with the regular hexagon representing the output limit range 23a, the vector extending to that intersection can be changed from the original voltage vectors 71 and 72 to a new voltage vector 73, thereby falling within the output limit range 23a.

In this modified example 1, even if it is determined that voltage vectors 71 and 72 obtained using the aforementioned equations (9) and (12) based on the conceptual example of the voltage command value correction algorithm shown in Figure 14 are outside the output limit range 23a, a voltage command value correction process (I) is performed to bring the voltage vectors within the output limit range 23a. Note that in this modified example 1 and the next modified example 2, the "range of output limit range 23a" includes each side and each vertex of the hexagon representing the range of output limit range 23a. Therefore, even if the coordinates representing the tip position of the modified voltage vector 73 overlap with a side or vertex of the hexagon, it is determined to be within the range of output limit range 23a. Note that this differs from the previously described configuration examples 1 to 3.

Just to be sure, the relationship between the αβ stationary coordinate system and the dq rotating coordinate system shown in FIG. 14 will be explained. While the _vα - _vβ plane of the αβ stationary coordinate system is fixed, the _vd - _vq plane of the dq rotating coordinate system rotates counterclockwise in synchronization with the rotational speed of the AC motor 50. The _vd -vq plane in the same figure represents the instant when the motor has rotated by the electrical angle _θre . Only the hexagon of _the output limit range 23a is shown on the _vα - _vβ plane. The voltage vectors 71, 72, and 73 other than the hexagon and the torque differential value line 93 are shown on the _vd - _vq plane and rotate, together with the _vd - _vq plane, in synchronization with the rotational speed of the AC motor 50.

The voltage command value correction process (I) shown in Figure 15 will now be described. This process is performed on the assumption that the torque differential value line 93 and the output voltage limit line have one or more intersections. The output voltage limit lines are the six sides that make up the hexagon representing the output limit range 23a. This process is stored in advance in the EEPROM or the like of the controllers 20a to 20c as a program that implements the functions of the voltage command value output unit 23 in this modified example 1 in software.

As shown in Figure 15, in the voltage command value correction process (I), step S201 first calculates the intersection of the torque differential value line 93 and the output voltage limit line. This process is similar to the "process of calculating the intersection of the six linear functions representing each side of the hexagon of the output limit range 23a and the line segment represented by the voltage vector 71" already described in configuration example 1. Therefore, a description of step S201 will be omitted.

If two or more solutions to the simultaneous equations are obtained by the intersection calculation process in step S201, there will be two or more intersections between the torque differential value line 93 and the output voltage limit line. In such a case (S203; Yes), the intersection selection process in the next step S205 is performed. On the other hand, if there are no multiple intersections between the two (S203; No), the process proceeds to the voltage command value change process in step S207. In this case, that intersection is selected.

In the intersection selection process of step S205, for example, of two or more sides with an intersection, the side that is closest to being parallel to the torque differential value line 93 is identified, and an intersection included in that side is selected. For example, in the example shown in FIG. 14, the side with intersection 80a is closer to being parallel to the torque differential value line 93 than the side with intersection 80b. Therefore, in this case, intersection 80a is selected from the sides with intersection 80a. Note that whether or not the sides of the hexagon and the torque differential value line 93 are close to being parallel is determined, for example, by calculating the difference between the slope of the side being judged and the slope of the torque differential value line 93, and determining that the smaller the difference between the slopes, the closer to being parallel.

In the voltage command value changing process of step S207, based on the coordinate information of the intersection selected in step S205 etc., vector information having the coordinates of the intersection as the tip coordinates is generated and replaced with the original voltage vector. For example, in the example shown in Fig. 14, information of a voltage vector 73 having the coordinates of the selected intersection 80a as the tip coordinates is generated based on the coordinate information of the selected intersection 80a, and the original voltage vectors 71 and 72 are replaced with this new voltage vector 73. As a result, the α-axis voltage command value v ^* _α and the β-axis voltage command value v ^* _β of the new voltage vector 73 are output from the voltage command value output unit 23.

In this manner, in the present modified example 1, when the side that is closest to parallel among the six sides constituting the hexagon representing _the output limit range 23a intersects with the torque differential value line 93 expressed by a predetermined function including the torque differential value τdot, the d-axis current i d , the q-axis current i _q and the angular velocity ω _re in the dq rotating coordinate system, the voltage vector extending to the intersection 80a is set as the voltage vector 73 (changed voltage vector), and the α-axis voltage command value v ^* _α and the β-axis voltage command value v ^* _β representing this vector are output as the αβ-axis voltage command values v ^** _α,β to the inverter 30 via the phase number conversion unit 24.

As a result, voltage vector 73 reaches one of the six sides of the hexagon representing output limit range 23a, and therefore voltage command value output unit 23 can output the maximum value as a feasible voltage command value to inverter 30 via phase number conversion unit 24 without using limiter control or the like that requires complex calculations and without performing integral calculations. Therefore, since no delay time associated with complex calculations or integral calculations occurs, it is possible to suppress a decrease in the response speed to torque differential command value τdot ^* and torque command value τ ^* .

[Modification 2 of Configuration Examples 1 to 3]
In the above-described first modification, the voltage command value correction process (I) executed as a function of the voltage command value output unit 23 has been described, but this process is based on the assumption that the torque differential value line 93 and the output voltage limit line have one or more intersections. Therefore, for example, as shown in FIG. 16 , if there is no intersection between the torque differential value line 94 and the output voltage limit line, that is, if the two do not intersect, the voltage command value correction process (I) cannot be applied.

The following describes voltage command value correction process (II) applicable when the torque differential value line 94 and the output voltage limit line do not intersect. As shown in FIG. 16, the torque differential value line 94 and the output voltage limit line do not intersect when the hexagon representing the output limit range 23a and the torque differential value line 94 are separated, or when the two are close to each other but the sides of the hexagon and the torque differential value line 93 are parallel or nearly parallel. This problem is resolved by voltage command value correction process (II) shown in FIG. 17. This process is stored in advance in the EEPROM or the like of controllers 20a-20c as a program that implements the functions of the voltage command value output unit 23 in this second modified example in software.

As shown in FIG. 17, in the voltage command value correction process (II), first, in step S301, the degree of parallelism between the torque differential value line 94 and the output voltage limit line (the six sides of the hexagon) is calculated. For example, the degree of parallelism of these lines is determined by calculating the difference in slope between the two lines, and determining that the smaller the difference in slope, the greater the degree of parallelism (closer to parallelism). Therefore, in step S301, a calculation process is performed to find the slope of the torque differential value line 94 and the slope of each side of the hexagon from a linear function that can identify these lines. As a result, six degrees of parallelism corresponding to each side of the hexagon are obtained.

In the next step S303, a specific limit line selection process, the side with the maximum parallelism is identified from the six parallelism degrees corresponding to the sides of the hexagon obtained in step S301. If there are two or more sides with the maximum parallelism, the side closest to the torque differential line 93 is identified from those two sides. For example, the separation distance is calculated from the positional relationship between the midpoint coordinates of those two sides and the coordinates of a predetermined position on the torque differential line 93.

In the subsequent process of obtaining coordinates of both ends in step S305, information on the coordinates of both ends of the output voltage limit line (specific side of the hexagon) identified in step S301 is obtained. For example, in the example shown in FIG. 16 , coordinate information on end points 81 and 82 is obtained. Then, in the process of calculating both voltage vectors in step S307, information on voltage vectors whose tip coordinates are the coordinates of the end points is generated based on this coordinate information, and the d-axis voltage _vd and the q-axis voltage _vq of both voltage vectors ( _vd , _vq ) are calculated. Then, by substituting the d-axis voltage _vd and the q-axis voltage _vq of these voltage vectors ( _vd , _vq ) into the aforementioned equation (7) related to the torque differential line 94, the torque differential value τdot of these voltage vectors is calculated.

In the endpoint selection process of step S309, the endpoint corresponding to the tip position of the larger of the torque differential values τdot of the two voltage vectors calculated in step S307 is selected. For example, in the example shown in Figure 16, of the two voltage vectors 74, 75, endpoint 82 of voltage vector 74, which has the larger torque differential value τdot, is selected.

Then, in the voltage command value changing process of step S311, the vector information of the voltage vector selected in step S309 is replaced with the original voltage vector. For example, in the example shown in Fig. 16, the selected voltage vector 74 is replaced with the original voltage vectors 71 and 72. As a result, the α-axis voltage command value v ^* _α and the β-axis voltage command value v ^* _β of the new voltage vector 74 are output from the voltage command value output unit 23.

In this manner, in the present modified example 2, when the side that is closest to parallel among the six sides constituting the hexagon representing the output limit range 23a with respect to the torque differential value line 94 expressed by a predetermined function including the torque differential value τdot, the d-axis current i _d , the q-axis current i _q and the angular velocity ω _re does not intersect with the torque differential value line 93 expressed by this predetermined function in the dq rotating coordinate system, the voltage vector 74 having the larger torque differential value τdot among the two voltage vectors 74, 75 representing the end points 81, 82 of the side that is closest to parallel is set as the (changed voltage vector), and the α-axis voltage command value v ^* _α and β-axis voltage command value v ^* _β representing this vector are output as αβ-axis voltage command values v ^** _α,β to the inverter 30 via the phase number conversion unit 24.

As a result, the voltage vector 74 reaches one of the six vertices of the hexagon representing the output limit range 23a, and the voltage command value output unit 23 can output the maximum value as a possible voltage command value to the inverter 30 via the phase number conversion unit 24 without using limiter control, which requires complex calculations, or performing integral calculations. Therefore, there is no delay time associated with complex calculations or integral calculations, and a decrease in the response speed to control commands can be suppressed.

[Modification 3 of Configuration Examples 1 to 3]
The above-described first and second modifications have different roles and are complementary to each other. That is, the first modification is based on the assumption that the torque differential line 93 and the output voltage limit line have one or more intersections, while the second modification is based on the assumption that there are no intersections between them. Therefore, it is better to provide a process for determining whether the torque differential line 93 or the like intersects with the output voltage limit line as preprocessing for the first and second modifications, and to configure an algorithm for allocating control so as to conform to the respective preconditions for the first and second modifications.

Furthermore, as explained in Configuration Example 3 with reference to FIG. 13, for example, after the AC motor 50 is started, the AC motor 50 may be controlled with the minimum voltage amplitude control of Configuration Example 1 until the output torque reaches the required torque, and after the output torque reaches the required torque, the AC motor 50 may be controlled by switching to the minimum current amplitude control of Configuration Example 3 to enable a smooth transition to MTPA control. In this way, an algorithm that links the minimum voltage amplitude control of Configuration Example 1 and the minimum current amplitude control of Configuration Example 3 depending on the operating conditions of the AC motor 50 is also useful.

In this modified example 3, the controller 20 (20a to 20c) of the motor system 10 (10a to 10c) is configured with an integrated control processing algorithm as shown in Figure 18, which enables coordination of minimum voltage amplitude control and minimum current amplitude control according to the operating conditions of the AC motor 50, and allocation of control that is compatible with the respective prerequisites for the modified examples 1 and 2 described above.

Since the functions of controllers 20a to 20c can be realized by the same hardware, such control switching can be handled by software. This integrated control process is stored in advance in the EEPROM of controller 20 as a program that realizes this process. Furthermore, this integrated control process is repeatedly executed at a predetermined control cycle by the CPU of controller 20 from the start to the end of operation control of AC motor 50.

As shown in FIG. 18 , in the integrated control process, first, a voltage vector calculation process (II) is performed in step S501. This process has been described with reference to FIG. 12 in the configuration example 3. Therefore, here, the controller 20c of the configuration example 3 causes the voltage information generation unit 21 to perform the following process in preparation for current amplitude minimum control. When the d-axis voltage command value v ^* _d is input from the voltage information generation unit 29, the voltage information generation unit 21 determines the d-axis voltage _vd , one of the two variables of the voltage vector ( _vd , _vq ), and therefore the remaining q-axis voltage _vq is also automatically determined. Therefore, a voltage vector 72 suitable for MTPA control is selected as the voltage command value v ^* _d,q and output to the coordinate conversion unit 22. The coordinate conversion unit 22 converts the input voltage command value v ^* _d,q from the dq rotating coordinate system to the αβ stationary coordinate system and outputs the αβ-axis voltage command value v ^* _α,β to the voltage command value output unit 23.

Then, in the next step S503, the voltage command value output unit 23 performs a process of determining whether or not the voltage vector 72 is within the output limit range 23a of the inverter 30. If this determination process determines that the voltage vector 72 is within the hexagonal range of the output limit range 23a (S503; Yes), there is no need to change the magnitude of the voltage vector 72, and therefore the voltage command value output unit 23 outputs the αβ-axis voltage command value v ^** _α,β to the phase number conversion unit 24. In this case, the current amplitude minimum control of configuration example 3 is performed.

In the following step S511, the phase number conversion unit 24 converts the input αβ-axis voltage command value v ^** _α,β from a two-phase voltage vector in the αβ stationary coordinate system into a three-phase voltage vector in the UVW stationary coordinate system, and outputs voltage command values v ^* _u,v,w . The voltage command values v ^* _u,v,w output from the phase number conversion unit 24 are input to the inverter 30. The inverter 30 performs a switching operation in accordance with the voltage command values v ^* _u,v,w, and the AC voltages vu _,v,w generated thereby are supplied to the AC motor 50.

On the other hand, if the determination process of step S503 determines that the voltage vector 72 is not within (outside) the hexagonal range of the output limit range 23a (S503; No), the voltage command value output unit 23 discards the voltage vector 72.

Then, in the integrated control process, the discarding of this voltage vector 72 triggers the voltage vector calculation process (I) to be performed in step S505. This process has been described with reference to FIG. 4A in the configuration example 1. Therefore, here, in preparation for performing voltage amplitude minimum control, the controller 20a of the configuration example 1 performs a process in which the voltage information generating unit 21 selects, as the voltage command values v ^* _d,q , the voltage vector 71 that minimizes the voltage amplitude for the input torque differential command value τdot ^* . The voltage command values v ^* _d,q of the selected vector 71 are converted into αβ-axis voltage command values v ^* _α,β by the coordinate conversion unit 22, and then output to the voltage command value output unit 23.

Then, in the next step S507, the voltage command value output unit 23 performs processing to determine whether or not the voltage vector 71 is within the output limit range 23a of the inverter 30. If it is determined that the voltage vector 71 is within the hexagonal range of the output limit range 23a (S507; Yes), there is no need to change the magnitude of the voltage vector 71, and therefore the voltage command value output unit 23 outputs the αβ-axis voltage command value v ^** _α,β to the phase number conversion unit 24. In this case, the voltage amplitude minimum control of Configuration Example 1 is performed. In the voltage command value output processing of step S511, processing similar to that described above is performed, and AC voltages v _u,v,w are supplied to the AC motor 50.

On the other hand, if the determination process of step S507 determines that the voltage vector 71 is not within the hexagonal range of the output limit range 23a (outside the range) (S507; No), further determination process is performed in step S509.

In step S509, a process is performed to determine whether the torque differential value line 91 and the output voltage limit line (the sides of the hexagon representing the output limit range 23a) intersect. This process is the "preprocessing for Modified Examples 1 and 2" described above. Therefore, if the voltage command value output unit 23 determines that they intersect (S509; Yes), the two will have one or more intersections, and the process proceeds to voltage command value correction process (I) of Modified Example 1, which assumes that there is an intersection (S200). On the other hand, if the voltage command value output unit 23 determines that the torque differential value line 91 and the output voltage limit line do not intersect (S509; No), the two will not intersect, and the process proceeds to voltage command value correction process (II) of Modified Example 2, which assumes that they do not intersect (there is no intersection) (S300).

The voltage command value correction process (I) (S200) of Modified Example 1 and the voltage command value correction process (II) (S300) of Modified Example 2 have already been described, so a detailed description will be omitted here. Even if voltage vector 71 or the like does not fit within the hexagon of output limit range 23a, these voltage command value correction processes (I) and (II) (S200, S300) change it to voltage vectors 73, 74, and 75 with dimensions on the sides and vertices of the hexagon (Figures 14 and 16). As a result, in such cases, so-called one-pulse control (single-pulse drive) is performed.

As a result of computer simulation of this voltage vector control in the integrated control processing by the controllers 20a to 20c, response characteristics such as output torque and voltage command values v ^* _u,v,w of the U, V, and W phases output to the inverter 30 are obtained as shown in Figures 19 and 20, and these will be described below. Note that the simulation conditions are different between Figures 19 and 20.

Simulation conditions for the results shown in FIG. 19: Rotational speed of AC motor 50: 1800 min ⁻¹ (constant)
Torque control unit control gain K: 5000 rad/sec
Torque command value τ ^* : Increases from 0 N·m to 1 N·m

Simulation conditions for the results shown in FIG. 20: Rotational speed of AC motor 50: 1000 min ⁻¹ (constant)
Torque control unit control gain K: 5000 rad/sec
Torque command value τ ^* : Increases from 0 N·m to 4 N·m

First, in the simulation results shown in FIG. 19, the output torque τ relative to the torque command value τ ^* rises relatively quickly ( FIG. 19(A) ). Furthermore, for a period of approximately 500 μsec from the start of control, the waveforms of the voltage command values v ^* _u,v,w for the U, V, and W phases remain stuck at the positive maximum voltage and the negative minimum voltage, indicating that one-pulse control was performed according to Modification Example 2 during this period ( FIG. 19(B) ). Since the torque command value τ ^* temporarily uses the overmodulation region, it is inferred that one-pulse control was performed according to Modification Example 2. Note that no overshoot occurred in the output torque τ (in the dashed ellipse shown in FIG. 19(A) ). Furthermore, it was confirmed that no noise or other artifacts were present in the waveforms of the output torque τ and the torque estimate value τhat even when switching control from Modification Example 2 to Configuration Example 1.

Note that Figure 25 shows the simulation results for the comparative example when current vector control was performed. The simulation conditions for this comparative example were set in the same way as the simulation conditions for the results shown in Figure 19, as follows. Therefore, the two can be compared.

The simulation conditions for the comparative example (FIGS. 24 and 25) are as follows:
Rotation speed of AC motor 50: 1800 min ⁻¹ (constant)
・ACR (current controller) response frequency: 5000 rad/sec
Torque command value τ ^* : Increases from 0 N·m to 1 N·m

19(A) and 25(A), it is clear that overshoot occurs in the output torque τ and the estimated torque value τhat in the current vector control of the comparative example (within the dashed ellipse in FIG. 25(A)). However, in the voltage vector control according to Modification Example 3 and Configuration Example 1, even if the αβ-axis voltage command value v ^** _α,β is limited by temporarily using the overmodulation region as described above, it can be confirmed from the dashed ellipse in FIG. 19(A) that no overshoot occurs in the output torque τ, etc.

Furthermore, as can be seen from the simulation results shown in Fig. 20, even when the simulation conditions are different, as in the simulation results of Fig. 19, even if the αβ-axis voltage command value v ^** _α,β is limited by temporarily using the overmodulation region, it can be confirmed within the dashed ellipse shown in Fig. 20(A) that no overshoot due to this occurs in the output torque τ, etc. Furthermore, from the simulation results shown in Figs. 19 and 20, it can be confirmed that a seamless transition occurs without the occurrence of noise, vibration, etc. when switching control from Modified Configuration 2 to Configuration Example 1.

In this way, in Modification Example 3, by performing integrated control processing in controller 20 (20a-20c), it is possible to link the minimum voltage amplitude control of Configuration Example 1 with the minimum current amplitude control of Configuration Example 3 depending on the operating conditions of AC motor 50, or to allocate control between Modification Examples 1 and 2 according to their respective preconditions. This makes it possible to switch between various algorithms (Configuration Examples 1-3, Modification Examples 1 and 2) related to this voltage vector control during operation control of AC motor 50. As a result, for example, AC motor 50 can be operated with the minimum voltage amplitude control of Configuration Example 1 from the time the AC motor 50 is started until the output torque reaches the required torque, and then switched to the minimum current amplitude control of Configuration Example 3 after the output torque reaches the required torque, thereby operating AC motor 50 with high power efficiency.

[Configuration example 4 of control system elements]
Next, a fourth configuration example (hereinafter referred to as "Configuration Example 4") of the control system elements of the motor system 10 (10d) will be described with reference to Figures 21 and 22. The controllers 20 (20a to 20c) constituting the motor systems 10 (10a to 10c) of Configuration Examples 1 to 3 described so far have not mentioned overcurrent protection of the current (motor current) flowing from the inverter 30 to the AC motor 50. However, overcurrent protection is an essential function in an actual machine. Therefore, the controller 20d of Configuration Example 4 employs a configuration that includes an overcurrent protection function. Note that the overcurrent protection function of Configuration Example 4 is used by adding it to, for example, Configuration Examples 1 to 3.

As shown in FIG. 21, in motor system 10d of configuration example 4, controller 20d has the functions of voltage information generator 21, coordinate converter 22, voltage command value output unit 23, phase number converter 24, subtractor 25', and torque controller 26'. In other words, motor system 10b of configuration example 2 (FIG. 6) has subtractor 25 and torque controller 26 in addition to voltage information generator 21, etc., so motor system 10d of configuration example 4 is configured in almost the same way as motor system 10b of configuration example 2.

However, in motor system 10d of configuration example 4, |i _lim | ² input from an external source and motor current |i| ² obtained from current sensors 41 to 43 are input to subtraction unit 25', which then calculates and outputs the current deviation (|i _lim | ² - |i| ² ) of motor current |i| ² from current limit value |i _lim | ^2. Here, current limit value |i _lim | ² and motor current |i| ² are expressed in square form, but this is simply to simplify the calculation process. Since it is necessary to calculate the square root of the sum of squares to calculate motor current |i| from motor currents iu _,v,w of U-, V-, and W-phases obtained from current sensors 41 to 43, subtraction unit 25' uses square form to simplify the calculation process by omitting the square root calculation.

The current deviation (|i _lim | ² - |i| ² ) output from the subtraction unit 25' is input to a torque control unit 26' having a control gain Ki. The torque control unit 26' multiplies the input current deviation (|i _lim | ² - |i| ² ) by the control gain Ki (Ki(|i _lim | ² - |i| ² )), and inputs this to the voltage information generation unit 21 instead of the torque differential command value τdot ^* .

However, the condition under which torque control unit 26' inputs τdot ^* = Ki(|i _lim | ² - |i| ² ) to voltage information generation unit 21 is when motor current |i| flowing through AC motor 50 reaches or exceeds current limit value i _lim (|i _lim | ² ≦ |i| ² ). Therefore, in any other case, that is, when motor current |i| is less than current limit value i _lim , torque differential command value τdot ^* is input to voltage information generation unit 21 in the same manner as in the configuration of controllers 20a to 20c of motor systems 10a to 10c.

21 shows the control system elements when the overcurrent protection function is activated. With the controller 20d configured in this manner, for example, if the current limit value i _lim is set to 8 A (amperes), the subtractor 25' detects that a current of 8 A or more flows through the AC motor 50, and this information is input to the torque control unit 26'. The torque control unit 26' inputs τdot ^* = Ki(|i _lim | ² - |i| ² ) to the voltage information generator 21, instead of the torque differential command value τdot ^* input from the outside. As a result, the voltage information generator 21 controls the motor current to be reduced to less than the current limit value i _lim , and the current flowing through the AC motor 50 is reduced to less than the current limit value i _lim .

As a result of computer simulation of the overcurrent protection function of the AC motor 50 using the controller 20d, the response characteristics of the output torque and the motor current |i| of the AC motor 50 shown in Figure 21 were obtained, which will now be explained. The simulation conditions were as follows:

Rotation speed of AC motor 50: 3600 min ⁻¹ (constant)
Control gain K of torque control section: 10 rad/sec
Torque command value τ ^* : Increased from 2 N·m to 6 N·m Current limit value i _lim : 8 A

In this simulation, it was known from the specifications of the AC motor 50 that a motor current |i| of 8 A or more would flow when the output torque reached 6 N·m or more. Therefore, the torque command value τ ^* was increased from 2 N·m to 6 N·m. At the start of control, the motor was operated at 2 N·m, and the motor current |i| was approximately 7 A, as can be seen from FIG. 22(C). When the torque command value τ ^* was subsequently increased to 6 N·m, the motor current |i| temporarily exceeded 8 A. However, the torque control unit 26' (current limiter) immediately functioned, and the operation control of the AC motor 50 switched from torque control mode to current amplitude control mode. In other words, it can be seen that the operation control was changed from torque control according to Configuration Example 1 to current amplitude control according to Configuration Example 4 (see FIG. 22(C)). In the current control mode, the motor current |i| was controlled and maintained at approximately 8 A, and therefore the output torque τ of the AC motor 50 was not controlled but maintained at approximately 3 N·m.

Configuration Example 4 has two limiters: a voltage limiter that controls the voltage vector within the output limit range 23a by the voltage command value output unit 23 originally provided in the controller 20, and a current limiter that functions by the subtraction unit 25' and torque control unit 26' described above. The simulation results shown in Figure 22 confirm that the transition from Configuration Example 1 to Configuration Example 4 is seamless and without any major shocks such as noise or vibrations occurring when switching control.

As described above, in configuration example 4, when the motor current |i| flowing through AC motor 50 is equal to or greater than (or exceeds) a predetermined current limit value (threshold value) i _lim , the torque control of AC motor 50 using this voltage vector control, which had been performed until then, is switched to current control of AC motor 50. This current control controls motor current |i| to be equal to or close to the predetermined current limit value. As a result, when motor current |i| equal to or greater than the predetermined current limit value i _lim flows through AC motor 50, motor current |i| is reduced to the predetermined current limit value, making it possible to protect AC motor 50 from breakdowns or failures caused by overcurrent.

In the above-described embodiment, the rotating coordinate system of the AC motor is described as having two axes, "a d-axis parallel to the magnetic pole direction of the rotor constituting the AC motor and a q-axis perpendicular to the d-axis," i.e., the rotating coordinate system is a dq rotating coordinate system. However, as long as it is a coordinate system of an AC motor, it may be a maximum torque control coordinate system or any other rotating coordinate system. It may also be a stationary coordinate system (fixed coordinate system). For example, as shown in FIG. 23(A), in a case where an arbitrary rotating coordinate system of γδ axes leads the dq rotating coordinate system (shown in light black in the figure) by a phase difference θ and rotates in synchronization with the rotor of the motor together with the dq rotating coordinate system, the following equation (17) can be obtained by substituting _vd and _vq in the rotating coordinate transformation equation expressed by the following equation (16) into the above-described equation (7) and rearranging:

Since this equation (17) is expressed as an equation of the torque differential value τdot using the voltages v _γ and v _δ of an arbitrary rotating coordinate system, it is possible to obtain the specified γ-axis voltage value v ^* _γ and the specified δ-axis voltage value v ^* _δ in the arbitrary rotating coordinate system by solving this equation (17). In other words, by having the voltage information generating unit perform calculation processing based on the inverse function of this equation (17), it is possible to generate rotating coordinate system voltage information corresponding to the two orthogonal axes of the arbitrary rotating coordinate system based on the torque differential value. Therefore, even when an arbitrary rotating coordinate system of an AC motor, including a maximum torque control coordinate system, is used, it is possible to perform torque response management type voltage vector control for a three-phase AC motor, as in the above-described embodiment.

Based on this concept, by setting θ = -θ _re as a special example of an arbitrary rotating coordinate system, it becomes possible to obtain an equation for the torque differential value τdot for the stationary coordinates (fixed coordinates) of the AC motor in the same way as for the arbitrary rotating coordinate system. θ _re is the electrical angle of the AC motor 50. By substituting v _d and v _q of the rotating coordinate transformation equation expressed by the following equation (18) into the above-mentioned equation (7) and rearranging, the following equation (19) is obtained.

By solving the following equation (19), it is possible to obtain the α-axis voltage command value v ^* _α and the β-axis voltage command value v ^* _β on the stationary coordinate system. From the above, it can be understood that the rotating coordinate system of the AC motor may be replaced with stationary coordinates (fixed coordinates). In other words, by having the voltage information generating unit perform calculation processing based on the inverse function of this equation (19), it is possible to generate stationary coordinate system voltage information corresponding to the two orthogonal axes of the stationary coordinate system based on the torque differential value. Therefore, even when the stationary coordinate system of the AC motor is used, torque response management type voltage vector control for a three-phase AC motor can be performed, as in the above-described embodiment.

In addition, in the above-described embodiment, a control device and control method for controlling an IPMSM AC motor 50 were described as an AC motor. However, synchronous AC motors, such as a surface permanent magnet synchronous motor (SPMSM) or a reluctance motor, can also be controlled using the above-described control device and control method. In addition, in the above-described embodiment, a case where a three-phase AC motor (motor 150) driven by a supply of three-phase AC power was controlled was described as an example. However, multi-phase AC motors, such as two-phase, five-phase, or seven-phase AC motors, can also be controlled using the above-described control device and control method.

In the case of a two-phase AC motor, since there is no need to perform phase number conversion, the phase number conversion unit 24 is not required, and the α-axis voltage command value v ^** _α and the β-axis voltage command value v ^** _β can be directly output from the voltage command value output unit 23 to the inverter 30. In the case of a five-phase AC motor or a seven-phase AC motor, a phase number conversion unit that converts the α-axis voltage command value v ^** _α and the β-axis voltage command value v ^** _β into five-phase or seven-phase voltage command values is required instead of the phase number conversion unit 24. In the case of a three-phase AC motor, the regular hexagonal output limit range 23a becomes a regular polygon with twice the number of vertices as the number of phases, such as a regular square for two phases, a regular decagon for five phases, and a regular tetragon for seven phases, depending on the number of phases of the motor.

Furthermore, in the above-described embodiment, the controller 20 is configured using a microcomputer board as a control device for an AC motor, but the controller 20 may also be configured using a logic circuit, gate array, PLD (Programmable Logic Device), or FPGA (Field Programmable Gate Array) that implements the functions of the voltage information generation unit 21, coordinate conversion unit 22, voltage command value output unit 23, phase number conversion unit 24, etc. in hardware.

Furthermore, in the above-described embodiment, an example has been described in which, instead of the torque (output torque) value generated on the output shaft of the AC motor, the torque estimate value τhat calculated by the controller 20 is input to the subtraction unit 25, and the difference between this and the torque command value τ ^* (τ ^* - τhat) is output from the subtraction unit 25 as an estimated deviation of the output torque τ. However, for example, a torque sensor capable of measuring the output torque may be attached to the shaft (output shaft) of the AC motor 50, and information on the output torque τ output from this torque sensor may be input to the subtraction unit 25. Since highly accurate information on the output torque τ can be input to the subtraction unit 25 in real time, it becomes possible to use real deviation information on the controlled object, rather than an estimated value, as the deviation of the output torque τ for control.

Although specific examples of the present invention have been described in detail above, these are merely examples and do not limit the scope of the claims. The technology described in the claims includes various modifications or variations of the above-mentioned specific examples. Furthermore, the technical elements described in this specification or drawings demonstrate technical utility either alone or in various combinations, and are not limited to the combinations described in the claims at the time of filing. Furthermore, the technology illustrated in this specification or drawings simultaneously achieves multiple objectives, and achieving one of these objectives is itself technically useful. Note that the descriptions in parentheses in the "Explanation of Symbols" column may clarify the correspondence between the terms used in the above-mentioned embodiments and the terms described in the claims.

10, 10a, 10b, 10c, 10d... Motor system 20, 20a, 20b, 20c, 20d... Controller (control device for AC motor)
21: Voltage information generating unit 22: Coordinate conversion unit 23: Voltage command value output unit 23a: Output limit range (range of output limit voltage)
24... Phase number conversion unit 25, 25', 27... Subtraction unit 26, 26a... Torque control unit 28... Current control unit 29... Voltage information generation unit 30... Inverter 31 to 36... IGBT
41 to 43... Current sensors 45... Rotation angle sensor 50... AC motor (AC electric motor)
51... Coordinate transformation element 52... Function element 53... Integral element 71, 71a, 71b, 71c, 72... Voltage vector 73 to 75... Voltage vector (changed voltage vector)
80a, 80b... intersection points 81, 82... end points 91 to 94... torque differential value lines (lines expressed by a predetermined function)
i _d ... d-axis current i ^* _d ... d-axis current command value i _q ... q-axis current i ^* _q ... q-axis current command value v _d ... d-axis voltage v ^* _d ... d-axis voltage command value (rotating coordinate system voltage information)
v _q ...q-axis voltage v ^* _q ...q-axis voltage command value (rotating coordinate system voltage information)
v _u,v,w ... AC voltage v ^* _u,v,w ... UVW phase voltage command value (voltage command value of AC voltage to be output to AC motor)
v ^* _α ... α-axis voltage command value (stationary coordinate system voltage information)
v ^** _α …α-axis voltage command value (voltage command value)
v ^* _β ... β-axis voltage command value (stationary coordinate system voltage information)
v ^** _β ... β-axis voltage command value (voltage command value)
τ...output torque τ ^* ...torque command value (output torque required for AC motor, control command value)
τdot: Torque differential value (time differential value of output torque required for AC motor)
τdot ^* ... Torque differential command value (torque differential value input as a command value, control command value)
θ _rm …Rotation angle (mechanical angle)
ω _re …angular velocity

Claims (7)

A control device for an AC motor that outputs a voltage command value of an AC voltage to be output to an AC motor to an inverter capable of supplying three-phase AC power to the AC motor,
The voltage command value is a time differential value of an output torque required of the AC motor, and is input as a command value and is generated based on a torque differential value expressed by the following equation:
a voltage information generating unit that generates rotating coordinate system voltage information corresponding to two orthogonal axes of a rotating coordinate system of the AC motor based on the torque differential value;
a coordinate conversion unit that converts the rotating coordinate system voltage information from the rotating coordinate system into stationary coordinate system voltage information corresponding to two orthogonal axes of a stationary coordinate system based on a rotation angle of an output shaft of the AC motor;
a voltage command value output unit that outputs the stationary coordinate system voltage information to the inverter as the voltage command value when a voltage vector represented by the stationary coordinate system voltage information is within a range of an output limit voltage of the inverter, and that outputs stationary coordinate system voltage information representing a changed voltage vector, which has been changed to a size that falls within the range of the output limit voltage, to the inverter as the voltage command value when the voltage vector is outside the range of the output limit voltage;
A control device for an AC motor, comprising:

However, A(i _d , i _q ), B(i _d , i _q ), and C(i _d , i _q , ω _re ) are
and
In the AC motor, two orthogonal axes of a rotating coordinate system are a d-axis parallel to the magnetic pole direction of a rotor constituting the AC motor and a q-axis orthogonal to the d-axis, and in the above equation,
τ with a dot "·" symbol is the torque differential value, i _d is the d-axis current, i _q is the q-axis current, ω _re is the electrical angular velocity of the output shaft of the AC motor, L _d is the inductance of the d-axis equivalent coil, L _q is the inductance of the q-axis equivalent coil, k _e is the back electromotive force constant, and R is the winding resistance of the coil of each phase that constitutes the AC motor. A control device for an AC motor that outputs a voltage command value of an AC voltage to be output to an AC motor to an inverter capable of supplying three-phase AC power to the AC motor,
the voltage command value is a time differential value of an output torque required of the AC motor, and is generated based on a torque differential value input as a command value;
a voltage information generating unit that generates rotating coordinate system voltage information corresponding to two orthogonal axes of a rotating coordinate system of the AC motor based on the torque differential value;
a coordinate conversion unit that converts the rotating coordinate system voltage information from the rotating coordinate system into stationary coordinate system voltage information corresponding to two orthogonal axes of a stationary coordinate system based on a rotation angle of an output shaft of the AC motor;
a voltage command value output unit that outputs the stationary coordinate system voltage information to the inverter as the voltage command value when a voltage vector represented by the stationary coordinate system voltage information is within a range of an output limit voltage of the inverter, and that outputs stationary coordinate system voltage information representing a changed voltage vector, which has been changed to a size that falls within the range of the output limit voltage, to the inverter as the voltage command value when the voltage vector is outside the range of the output limit voltage;
It is equipped with
When the two axes of the rotating coordinate system are a d-axis parallel to the magnetic pole direction of a rotor constituting the AC motor and a q-axis perpendicular to the d-axis,
The voltage information generating unit, in the rotating coordinate system,
a control device for an AC motor, characterized in that it generates, as the rotating coordinate system voltage information, coordinate information of a voltage vector having a magnitude up to an intersection point that intersects with a straight line expressed by a predetermined function including "the torque differential value, a d-axis current, a q-axis current, and the angular velocity of the output shaft" at the shortest distance. A control device for an AC motor that outputs a voltage command value of an AC voltage to be output to an AC motor to an inverter capable of supplying three-phase AC power to the AC motor,
the voltage command value is a time differential value of an output torque required of the AC motor, and is generated based on a torque differential value input as a command value;
a voltage information generating unit that generates rotating coordinate system voltage information corresponding to two orthogonal axes of a rotating coordinate system of the AC motor based on the torque differential value;
a coordinate conversion unit that converts the rotating coordinate system voltage information from the rotating coordinate system into stationary coordinate system voltage information corresponding to two orthogonal axes of a stationary coordinate system based on a rotation angle of an output shaft of the AC motor;
a voltage command value output unit that outputs the stationary coordinate system voltage information to the inverter as the voltage command value when a voltage vector represented by the stationary coordinate system voltage information is within a range of an output limit voltage of the inverter, and that outputs stationary coordinate system voltage information representing a changed voltage vector, which has been changed to a size that falls within the range of the output limit voltage, to the inverter as the voltage command value when the voltage vector is outside the range of the output limit voltage;
It is equipped with
When the two axes of the rotating coordinate system are a d-axis parallel to the magnetic pole direction of a rotor constituting the AC motor and a q-axis perpendicular to the d-axis,
the voltage information generating unit generates the rotating coordinate system voltage information based on the torque differential value, a d-axis current, a q-axis current, an angular velocity of the output shaft, and a time differential command value of the d-axis current in MTPA (Maximum Torque Per Ampere) control. A control device for an AC motor that outputs a voltage command value of an AC voltage to be output to an AC motor to an inverter capable of supplying three-phase AC power to the AC motor,
the voltage command value is a time differential value of an output torque required of the AC motor, and is generated based on a torque differential value input as a command value;
a voltage information generating unit that generates rotating coordinate system voltage information corresponding to two orthogonal axes of a rotating coordinate system of the AC motor based on the torque differential value;
a coordinate conversion unit that converts the rotating coordinate system voltage information from the rotating coordinate system into stationary coordinate system voltage information corresponding to two orthogonal axes of a stationary coordinate system based on a rotation angle of an output shaft of the AC motor;
a voltage command value output unit that outputs the stationary coordinate system voltage information to the inverter as the voltage command value when a voltage vector represented by the stationary coordinate system voltage information is within a range of an output limit voltage of the inverter, and that outputs stationary coordinate system voltage information representing a changed voltage vector, which has been changed to a size that falls within the range of the output limit voltage, to the inverter as the voltage command value when the voltage vector is outside the range of the output limit voltage;
It is equipped with
When the voltage vector is outside the range of the output limit voltage, the voltage command value output unit
a torque differential value being reduced until the changed voltage vector falls within a range of the output limit voltage. A control device for an AC motor that outputs a voltage command value of an AC voltage to be output to an AC motor to an inverter capable of supplying three-phase AC power to the AC motor,
the voltage command value is a time differential value of an output torque required of the AC motor, and is generated based on a torque differential value input as a command value;
a voltage information generating unit that generates rotating coordinate system voltage information corresponding to two orthogonal axes of a rotating coordinate system of the AC motor based on the torque differential value;
a coordinate conversion unit that converts the rotating coordinate system voltage information from the rotating coordinate system into stationary coordinate system voltage information corresponding to two orthogonal axes of a stationary coordinate system based on a rotation angle of an output shaft of the AC motor;
a voltage command value output unit that outputs the stationary coordinate system voltage information to the inverter as the voltage command value when a voltage vector represented by the stationary coordinate system voltage information is within a range of an output limit voltage of the inverter, and that outputs stationary coordinate system voltage information representing a changed voltage vector, which has been changed to a size that falls within the range of the output limit voltage, to the inverter as the voltage command value when the voltage vector is outside the range of the output limit voltage;
It is equipped with
When the two axes of the rotating coordinate system are a d-axis parallel to the magnetic pole direction of a rotor constituting the AC motor and a q-axis perpendicular to the d-axis,
When the voltage vector is outside the range of the output limit voltage, the voltage command value output unit calculates, in the rotating coordinate system,
a voltage vector extending to a point of intersection of a line expressed by a predetermined function including "the torque differential value, a d-axis current, a q-axis current, and the angular velocity of the output shaft" and a side that is closest to parallel among six sides constituting a hexagon that represents the range of the output limit voltage, is set to the changed voltage vector, and static coordinate system voltage information that represents the changed voltage vector is output to the inverter as the voltage command value. A control device for an AC motor that outputs a voltage command value of an AC voltage to be output to an AC motor to an inverter capable of supplying three-phase AC power to the AC motor,
the voltage command value is a time differential value of an output torque required of the AC motor, and is generated based on a torque differential value input as a command value;
a voltage information generating unit that generates rotating coordinate system voltage information corresponding to two orthogonal axes of a rotating coordinate system of the AC motor based on the torque differential value;
a coordinate conversion unit that converts the rotating coordinate system voltage information from the rotating coordinate system into stationary coordinate system voltage information corresponding to two orthogonal axes of a stationary coordinate system based on a rotation angle of an output shaft of the AC motor;
a voltage command value output unit that outputs the stationary coordinate system voltage information to the inverter as the voltage command value when a voltage vector represented by the stationary coordinate system voltage information is within a range of an output limit voltage of the inverter, and that outputs stationary coordinate system voltage information representing a changed voltage vector, which has been changed to a size that falls within the range of the output limit voltage, to the inverter as the voltage command value when the voltage vector is outside the range of the output limit voltage;
It is equipped with
When the two axes of the rotating coordinate system are a d-axis parallel to the magnetic pole direction of a rotor constituting the AC motor and a q-axis perpendicular to the d-axis,
When the voltage vector is outside the range of the output limit voltage, the voltage command value output unit calculates, in the rotating coordinate system,
a control device for an AC motor, characterized in that, when the most parallel side of six sides constituting a hexagon representing the range of the output limit voltage does not intersect with the line expressed by a predetermined function including "the torque differential value, a d-axis current, a q-axis current, and the angular velocity of the output shaft," the voltage vector representing both end points of the most parallel side, which has a larger torque differential value, is set as the changed voltage vector, and stationary coordinate system voltage information representing the changed voltage vector is output to the inverter as the voltage command value. 1. A method for controlling an AC motor, comprising: outputting a voltage command value of an AC voltage to be output to an AC motor to an inverter capable of supplying three-phase AC power to the AC motor,
the voltage command value is a time differential value of an output torque required of the AC motor, and is input as a command value and is generated based on a torque differential value expressed by the following equation:

However, A(i _d , i _q ), B(i _d , i _q ), and C(i _d , i _q , ω _re ) are
and
In the AC motor, two orthogonal axes of a rotating coordinate system are a d-axis parallel to the magnetic pole direction of a rotor constituting the AC motor and a q-axis orthogonal to the d-axis, and in the above equation,
τ with a dot "·" symbol is the torque differential value, i _d is the d-axis current, i _q is the q-axis current, ω _re is the electrical angular velocity of the output shaft of the AC motor, L _d is the inductance of the d-axis equivalent coil, L _q is the inductance of the q-axis equivalent coil, k _e is the back electromotive force constant, and R is the winding resistance of the coil of each phase that constitutes the AC motor.