JP6541407B2 - Motor control device and program - Google Patents

Description

  The present invention relates to a motor control device and program capable of precisely controlling a motor.

  In order to conduct a characteristic test of a vehicle, it is conceivable to actually run the vehicle on a test course as one method. However, securing a test course every time of a characteristic test and transporting a car to a test course are bothersome. In addition, the characteristic test needs to be performed under various conditions such as a dry road surface, a wet road surface, a freeze road and the like. It was very difficult to set the condition of the test course to the desired condition, and it took too long to wait until the weather conditions were established, and it was difficult to properly and quickly perform the characteristic test.

  In order to cope with such a problem, for example, as shown in Patent Document 1, there is proposed a test apparatus which simulates a force received by a vehicle from a road surface or the like and performs a characteristic test of the vehicle in a laboratory. When conducting a characteristic test using this type of test apparatus, the wheel is removed from the 4-axis hub of the automobile, and four absorbing motors are mounted in place of the wheel. These absorption motors are driven by an inverter. The torque calculated to be applied to the 4-axis hub when the car is actually run on the test course is calculated by the computer, and the calculated torque is reproduced by the absorbing motor, and the car is actually run on the test course It is thought that the same phenomenon can be reproduced.

JP 2012-135159 A

However, when the traveling characteristics obtained by the above-described test apparatus and the actual traveling characteristics on the test course are compared, a slight difference occurs between the two. For example, when a test that causes vibrations in a car is conducted, the vibration condition in the test apparatus is different from the vibration frequency in the test device compared to the vibration condition due to the actual running characteristics. Be As a result of investigating the cause of such a difference, it was found that a delay was caused in the torque control of the absorbing motor. That is, after the torque command is given to the inverter, the torque command is reflected on the absorbing motor with a dead time of several milliseconds and a first-order delay whose time constant is several milliseconds. It is considered that these results in differences in the measurement results.
The present invention has been made in view of the above-described circumstances, and it is an object of the present invention to provide a motor control device and program capable of precisely controlling a motor (that is, compensating for a delay).

In order to solve the above-mentioned subject, a motor control device of the present invention is a motor control device which supplies the torque command value to an inverter which applies a drive voltage to a motor based on a supplied torque command value, A future torque calculation unit that calculates a future torque estimated value corresponding to a torque estimated to be realized by the motor in the future, and a torque command value output unit that outputs the torque command value based on the future torque estimated value And a load model unit that models a load connected to the motor into a model including a first element and outputs an estimated angular velocity value of the first element, the future torque calculation unit including: The future torque estimated value is output based on the angular velocity estimated value of the motor and the angular velocity estimated value of the first element .

  According to the motor control device and program of the present invention, the motor can be precisely controlled.

1 is a block diagram of a vehicle test system according to an embodiment of the present invention. It is a schematic diagram of a physical model adopted in this embodiment. It is a block diagram which shows the whole structure of the algorithm of this embodiment. It is a block diagram showing composition of an algorithm of a future torque operation part. (A) is a gain diagram of the torque transfer characteristic of an inverter and an absorption motor, (b) is the phase diagram. It is a block diagram which shows the structure of the algorithm of a comparative example. (A) is a gain diagram of the torque transfer characteristic seen from the engine of a motor vehicle, (b) is the phase diagram. (A) is the measurement result of the rotational speed of the drive wheel of a motor vehicle, (b) is the measurement result of the axial torque. It is the figure which expanded the time-axis of Fig.8 (a), (b).

Configuration of Embodiment
(Hardware configuration)
First, the hardware configuration of a vehicle test system according to an embodiment of the present invention will be described with reference to the block diagram shown in FIG.
In FIG. 1, an automobile 40 includes an engine 41, a transmission 42, and the like, and is completed to be able to actually travel on a road surface. The automobile 40 may be any of front wheel drive, rear wheel drive, and four wheel drive. However, in the automobile 40, the wheels are removed from the four-shaft hubs of the front left wheel, the front right wheel, the rear left wheel, and the rear right wheel, and the tire type bearings 50FL, 50FR, 50RL, 50RR are mounted. Hereinafter, these four tire-type bearings 50FL, 50FR, 50RL, and 50RR are collectively referred to as "tire-type bearings 50". The tire-type bearing 50 has an outline similar to the actual wheel and tire. However, the automobile 40 is supported in a floating state at a predetermined distance from the floor surface so that the tire type bearing 50 does not contact the floor surface.

  The rotary shafts of absorption motors 60FL, 60FR, 60RL, and 60RR, which are induction motors, are coupled to the tire-type bearings 50, respectively. Further, four sensor units 52FL, 52FR, 52RL, 52RR for detecting the rotational speed and axial torque of the rotary shaft are attached to the rotary shaft of these absorption motors. Further, the four inverters 62FL, 62FR, 62RL, 62RR supply an AC voltage PWM-modulated to the absorbing motors 60FL, 60FR, 60RL, 60RR. The four absorption motors 60FL, 60FR, 60RL, and 60RR described above are collectively called “absorption motor 60”, and the four sensor units 52FL, 52FR, 52RL, and 52RR are collectively called “sensor unit 52”. The four inverters 62FL, 62FR, 62RL, 62RR are collectively referred to as "inverter 62".

  The control device 30 is, for example, a general computer, a CPU (Central Processing Unit) 32, a ROM (Read Only Memory) 33 storing an IPL (Initial Program Loader), etc., and a RAM (RAM) used as a work memory of the CPU 32. A random access memory 34, an operating system (OS), a hard disk drive (HDD) 35 storing application programs and various data, and a bus 36 interconnecting them are provided.

  The OS and application programs are expanded in the RAM 34 and executed by the CPU 32. Further, under the control of the CPU 32, the control device 30 receives (inputs) an output interface 37 for outputting a torque command signal to each inverter 62, and receives (inputs) the axial torque and rotational speed of the four axes detected by the sensor unit 52. And an input interface 38 for outputting to 36.

(Physical model)
Next, the physical model adopted in the present embodiment will be described with reference to the schematic view shown in FIG.
FIG. 2 is a model diagram of a linear two-inertia model, where element 1 has inertia J ₁ and element 2 has inertia J ₂ . The angular velocities of the elements 1 and 2 are ω ₁ and ω ₂ , and the torques are τ ₁ and τ ₂ . There is a relationship of τ ₁ = J ₁ · dω ₁ / dt and τ ₂ = J ₂ · dω ₂ / dt among these numerical values. The elements 1 and 2 are connected by a transmission system of a spring constant K and a viscous damping coefficient C.

  “Element 2” in FIG. 2 corresponds to a four-axis hub of the automobile 40. In the configuration shown in FIG. 1, since the tire-type bearing 50 is mounted on the 4-axis hub, the element 2 may be considered to correspond to the tire-type bearing 50. The “element 1” corresponds to various elements of the car 40 exerting a force on the four-axis hub of the car 40. It is assumed that the tire type bearing 50 is removed from the automobile 40 and the actual wheel and tire are mounted to run on the road surface.

  The tire is elastic because it is made of rubber as a main component, and is distorted in the rotational direction, so that the torque exerted on the hub from the rotational state of the tire can be calculated by the calculation of the spring using the Hooke's law. Also, the tire is slipping and, depending on the slip characteristics, exerts more complex forces on the hub. Furthermore, the movement of the suspension (not shown) of the motor vehicle 40 also exerts a force on the hub. In FIG. 2, these various elements are put together and modeled as "element 1".

(algorithm)
Next, with reference to the block diagram shown in FIG. 3, the overall configuration of the algorithm in the present embodiment will be described.
This algorithm is a block diagram showing the function of the application program executed by the control device 30. The control device 30 outputs a torque command value τ _v (t) to be generated in the four-axis tire type bearing 50 to the inverter 62 at time t. Although FIG. 3 shows one value, such as “torque command value τ _v (t)”, it is actually an independent value for each of the four axes. The same is true for various other values shown in FIG.

The torque command value τ _v (t) is not immediately reflected as the torque of the absorbing motor 60, but is reflected with the dead time T _b and the first delay of the time constant T _c . Therefore, the inverter 62 and the absorbing motor 60 have a transfer function “exp (−T _b s) / (1 + T _c s)”. The dead time T _b and the time constant T _c can be determined by measuring the characteristics of the inverter 62 and the absorption motor 60.

A future torque estimated value τ _e2 (t + Δt), which is an estimated value of the torque τ ₂ (see FIG. 2) at a future time (t + Δt) from the current time t by the time difference Δt from the current time t, is input. The method of calculating the future torque estimated value τ _e2 (t + Δt) will be described later. Emulation unit 14 _applies a transfer function “exp (−Te _b · s) to input future torque estimated value τ _e2 (t + Δt) based on _dead time T _eb set by the user and time constant T _ec. ) / (1 + T _ec s) ”, and the result is output as a current torque estimated value τ _e2 (t).

The dead time T _eb and the time constant T _ec are values that can be arbitrarily set by the user, but in practice, they are substantially equal to the measurement results of the dead time T _b and the time constant T _c of the inverter 62 and the absorption motor 60 Set to Thus, the emulation unit 14 has a function of emulating the characteristics of the inverter 62 and the absorption motor 60. The above-mentioned time difference Δt is a value of time delay determined based on the _dead time T _eb and the time constant T _ec .

The wheel speed calculation unit 16 also receives an estimated current torque value τ _e2 (t) and a measured torque value τ _s2 (t) which is a measured value of the torque τ ₂ measured by the sensor unit 52. Here, the difference between the current torque estimated value τ _e2 (t) and the torque measured value τ _s2 (t) will be described. The current torque estimated value τ _e2 (t) is a value calculated internally in the CPU 32, and if 64-bit double precision floating point is applied, it has approximately 16 significant digits in decimal conversion. On the other hand, the effective digit in the control device 30 of the torque measurement value τ _s2 (t) is the same as the current torque estimated value τ _e2 (t), but the substantial effective digit is the accuracy of the sensor unit 52 and the input interface 38 Depends on the resolution of

Even if highly accurate ones are used as the sensor unit 52 and the input interface 38, the effective effective digit of the torque measurement value τ _s2 (t) is smaller than that of the current torque estimation value τ _e2 (t). Therefore, if the current torque estimated value τ _e2 (t) and the torque measured value τ _s2 (t) substantially match (more exactly, if the difference between the two is less than or equal to a predetermined difference threshold Δτ _th) It is considered desirable to use the value of the current torque estimated value τ _e2 (t) with many significant digits as it is. Therefore, in such a case, the wheel speed calculation unit 16 divides the current torque estimated value τ _e2 (t) by the inertia J ₂ (see FIG. 2), and integrates the division result (time integration) to estimate the angular velocity ω ₂ An angular velocity estimated value ω _e2 (t) which is a value is output.

On the other hand, assuming that the difference between τ _e2 (t) and τ _s2 (t) exceeds the difference threshold Δτ _th , this difference is the error or resolution of the sensor unit 52, and the torque actually generated by the absorbing motor 60. It is considered that the error or resolution of That is, due to these errors and resolution, the torque generated by the absorbing motor 60 deviates from the torque command value τ _v (t), and the amount of this deviation is the difference between τ _e2 (t) and τ _s2 (t) It is considered to have appeared as Therefore, in such a case, the estimated angular velocity value ω _e2 (t) is corrected to be slightly increased or decreased. That is, feedback control is performed via the angular velocity estimated value ω _e2 (t) so that the current torque estimated value τ _e2 (t) _{approaches the} torque measured value τ _s2 (t).

The rotation compensation controller 18 receives the estimated angular velocity value ω _e2 (t) and the measured angular velocity value ω _s2 (t) which is the measured value of the angular velocity ω ₂ (output value of the velocity sensor). Assuming that there is a difference between the two, it is considered that this is due to an error or resolution of torque actually generated by the absorbing motor 60. That is, the torque generated by the absorbing motor 60 deviates from the torque command value τ _v (t) due to the influence of an error or resolution, and the amount of this deviation is finally the difference between ω _e2 (t) and ω _s2 (t) It is considered to have appeared as Therefore, the rotation compensation controller 18 outputs a torque correction value Δτ (t) based on this difference. Thus, the estimated angular velocity value ω _e2 (t) is feedback-controlled to be asymptotically _{approached to the} measured angular velocity value ω _s2 (t).

The vehicle model unit 10 (load model unit) emulates the behavior of the automobile 40, and estimates the angular velocity ω ₁ (see FIG. 2) by sequentially inputting the current torque estimated value τ _e2 (t). The angular velocity estimated value ω _e1 (t) which is a value is sequentially output. The future torque calculator 12 outputs the above-mentioned future torque estimated value τ _e2 (t + Δt) based on the estimated angular velocity values ω _e1 (t) and ω _e2 (t) of the elements 1 and 2 at the current time t.

Here, the details of the future torque calculation unit 12 will be described with reference to the block diagram shown in FIG.
Subtracter 70 in FIG. 4 subtracts the estimated angular velocity value ω _e1 (t) from the angular velocity estimate ω _e2 (t), and outputs the result as an angular velocity difference estimates Δω _e (t). The proportional unit 71 multiplies the angular velocity difference estimated value Δω _e (t) by a predetermined value P. The predetermined value P is a value corresponding to the above-described time difference Δt. Thereby, the proportional unit 71 outputs an estimated angle difference Δθ which is an estimated value of the difference between the rotation angles which will appear in the elements 1 and 2 in the future by the time difference Δt on the basis of the rotation angle at the current time t.

Differentiator 72, by the estimated angle difference Δθ is differentiated (time differential), the difference only with the difference in angular velocity estimate in future Δω _e (t + Δt) and the angular velocity difference estimates [Delta] [omega _e of the current time t (t) the time difference Delta] t Output It should be noted that, since differentiating the estimated angle difference Δθ in practice may make the operation of the future torque calculation unit 12 unstable, the differentiator 72 desirably performs approximate differentiation. The adder 73 adds the difference output from the differentiator 72 and the angular velocity difference estimated value Δω _e (t) at the current time t to obtain the angular velocity difference estimated value Δω _e (t + Δt) in the future by the time difference Δt. Output.

The spring calculation unit 80 simulates the behavior of the spring constant K and the viscous damping coefficient C shown in FIG. 2 based on Hooke's law. The integrator 81 integrates the angular velocity difference estimated value Δω _e (t + Δt), and the proportional unit 82 multiplies the integration result by the spring constant K. The proportional unit 84 multiplies the viscous damping coefficient C by the angular velocity difference estimated value Δω _e (t + Δt). The adder 83 adds the multiplication results of the proportional units 82 and 84, and outputs the result as the future future torque estimated value τ _e2 (t + Δt) by the time difference Δt.

Returning to FIG. 3, the future torque estimated value τ _e2 (t + Δt) output from the future torque calculation unit 12 is supplied to the above-described emulation unit 14 and also to the adder 20. In the adder 20, the future torque estimated value τ _e2 (t + Δt) and the torque correction value Δτ (t) are added, and the addition result is supplied to the inverter 62 as a torque command value τ _v (t). Thus, in the present embodiment, the torque command value τ _v (t) is calculated based on the future torque estimated value τ _e2 (t + Δt) corresponding to the torque estimated to be realized in the future by the absorbing motor 60. In order to obtain it and supply it to the inverter 62, it is possible to compensate the dead time _Tb and the first-order delay of the time constant _Tc , and it is possible to control the absorbing motor 60 precisely.

(How to determine the time difference Δt)
Next, a method of determining the above-described time difference Δt will be described.
In setting the time difference Δt, it is preferable to measure the transfer function of the torque measurement value τ _s2 (t) with respect to the torque command value τ _v (t), that is, the transfer characteristics of the inverter 62 and the absorbing motor 60. At that time, although it is ideal that the tire type bearing 50 is not attached to the absorbing motor 60, it may be connected to the automobile 40 through the tire type bearing 50.

When a test signal such as a sine wave or a sweep signal is supplied to the inverter 62 as a torque command value τ _v (t), a torque measurement value τ _s2 (t) corresponding to the test signal is output from the sensor unit 52, thereby gain Diagrams and phase diagrams can be drawn. FIG. 5 (a) is a gain diagram based on the measurement of the torque measurement value τ _s2 (t) with respect to the torque command value τ _v (t), and FIG. 5 (b) is a phase diagram thereof. In FIG. 5 (b), when the frequency of the torque command value τ _v (t) is 100 [Hz], the phase delay occurring in the torque measurement value τ _s2 (t) is as shown by the one-dot chain line in the figure. It is about 52 [deg].

  When this phase delay is converted to time, it becomes (52 [deg] / 360 [deg]) / 100 [Hz] = 1.44 [ms]. Therefore, the time difference Δt may be set to 1.44 [ms]. However, the time difference Δt may be appropriately increased or decreased from 1.44 [ms] so as to actually obtain a good response. Further, since the characteristics in FIGS. 5A and 5B are the characteristics actually measured for the specific inverter 62 and the absorption motor 60, when the types of the inverter 62 and the absorption motor 60 change, the desirable range of the time difference Δt is Of course it also fluctuates.

[Comparative example]
Next, in order to clarify the effect of the above-mentioned embodiment further, the contents of a comparative example are explained. The hardware configuration of this comparative example is similar to that of the above embodiment (FIG. 1), but the contents of the algorithm executed in the control device 30 are different. The algorithm in the present comparative example will be described with reference to the block diagram shown in FIG. In addition, in FIG. 6, the same code | symbol is attached | subjected to the part corresponding to each part of FIGS. Further, in the present comparative example, although the linear two-inertia model (FIG. 2) is not assumed, estimated values of the rotational speed of the wheel and the axial torque of the wheel are obtained. Since these correspond to the element 2 in FIG. 2, in FIG. 6, the signals corresponding to the signals shown in FIG. 3 have the same signal names.

6, the emulation unit 14 in FIG. 3 is not provided, the vehicle model unit 110 is provided instead of the vehicle model unit 10, and the torque calculation unit 112 is provided instead of the future torque calculation unit 12. . The vehicle model unit 110 emulates the behavior of the automobile 40, and is a value obtained by obtaining the angular velocity of the wheel based on the vehicle model by sequentially inputting the current torque estimated value τ _e2 (t). The angular velocity model estimated value ω _m2 (t) is sequentially output.

The torque calculator 112 receives the angular velocity model estimated value ω _m2 (t) output from the vehicle model unit 110 and the angular velocity estimated value ω _e2 (t) output from the wheel velocity calculator 16. The angular velocity model estimated value ω _m2 (t) is obtained by emulating the behavior of the automobile 40, and the angular velocity estimated value ω _e2 (t) is the current torque estimated value τ _e2 (t). It is simply obtained by dividing by and integrating the result of division. Therefore, it is usually considered that the angular velocity model estimated value ω _m2 (t) more accurately represents the angular velocity of the wheel.

On the other hand, since the estimated angular velocity value ω _e2 (t) is obtained by the wheel velocity calculation unit 16 while comparing the current estimated torque value τ _e2 (t) with the measured torque value τ _s2 (t) There is no large deviation from the actual angular velocity of the absorbing motor 60. Therefore, when the difference between the angular velocity model estimated value ω _m2 (t) and the angular velocity estimated value ω _e2 (t) is equal to or smaller than a predetermined angular velocity threshold Δω _th , the torque calculation unit 112 calculates the angular velocity model estimated value ω _m2 ( Calculate the current torque estimate τ _e2 (t) by temporally differentiating t) and multiplying by the inertia of the wheel.

On the other hand, when the difference between the angular velocity model estimated value ω _m2 (t) and the angular velocity estimated value ω _e2 (t) exceeds the angular velocity threshold Δω _th , the angular velocity estimated value ω _e2 (t) is time differentiated and the inertia of the wheel is The current torque estimate τ _e2 (t) is calculated by multiplying Then, the current torque estimated value τ _e2 (t) and the torque correction value Δτ (t) are added in the adder 20, and the addition result is output to the inverter 62 as the torque command value τ _v (t).

[Effect of the embodiment]
Next, the effects of the present embodiment will be described in detail.
7 (a) and 7 (b) show torque transfer characteristics as viewed from the engine 41 of the automobile 40, FIG. 7 (a) is a gain diagram, and FIG. 7 (b) is a phase diagram. The thick broken lines in FIGS. 7A and 7B indicate the actual traveling characteristics when the automobile 40 is traveled on the test course. The thin solid line is the characteristic according to the present embodiment (FIG. 3), and the thin broken line is the characteristic according to the comparative example (FIG. 6). It is understood that the characteristics of the present embodiment almost coincide with the actual traveling characteristics. On the other hand, in the comparative example, it can be seen that the characteristics particularly around 4 to 6 Hz slightly deviate from the actual traveling characteristics.

  Next, with reference to FIGS. 8 (a) and 8 (b), test results when chip-in / chip-out is performed in the automobile 40 will be described. Note that tip-in and tip-out are steps of changing the degree of opening of the accelerator. FIG. 8 (a) shows the measurement result of the rotational speed of the drive wheel of the automobile 40, and FIG. 8 (b) shows the measurement result of the shaft torque. The period Tx shown in FIG. 8A is “one second”. According to the actual running characteristic shown by the thick broken line, when tip-in / tip-out is performed, damped vibration of about 6 Hz occurs. Further, it can be seen that the characteristics of the present embodiment shown by the thin solid line almost overlap the actual traveling characteristics.

  Further, according to the characteristics of the comparative example indicated by the thin broken line, although the damping vibration certainly occurs, the frequency is lowered to about 5.5 Hz as compared with the actual traveling characteristics. Moreover, it can be understood that the convergence of the vibration is faster than the actual traveling characteristic. Next, FIGS. 9A and 9B show enlarged views in which the time axis in the vicinity of the period Tx in FIGS. 8A and 8B is enlarged. Although the time axis is enlarged, it is understood that the characteristics of the present embodiment indicated by the thin solid line substantially match the actual traveling characteristics indicated by the thick broken line.

[Summary of composition and effect]
As described above, the motor control device (30) applied to the present embodiment is an inverter (62) that applies a drive voltage to the motor (60) based on the supplied torque command value (τ _v (t)). respect, the a torque command value (tau _v (t)) a motor control device for supplying (30), the future torque estimated value corresponding to the torque estimated to be realized in the future by the motor ( and a torque command value for outputting the torque command value (τ _v (t)) based on the future torque estimated value (τ _e2 (t + Δt)) for computing a future torque computing unit (12) for computing τ _e2 (t + Δt) And an output unit (20).

By outputting the torque command value (τ _v (t)) based on the future torque estimated value (τ _e2 (t + Δt)), the delay in the motor (60) and the inverter (62) can be compensated, 60) can be precisely controlled.

The motor control device (30) emulates the operation of the motor (60) and the inverter (62), and based on the future torque estimated value (τ _e2 (t + Δt)), the motor control device (30) The present invention is further characterized by further including an emulation unit (14) that outputs a current torque estimated value (τ _e2 (t)). Thereby, in addition to the future torque estimated value (τ _e2 (t + Δt)), the current torque estimated value (τ _e2 (t)) can also be acquired.

Further, the motor control device (30) is based on the current torque estimated value (τ _e2 (t)) and a torque measurement value (τ _s2 (t)) which is a result of measuring the torque of the motor (60). It further comprises a velocity calculating unit (16) for outputting an estimated angular velocity value (ω _e2 (t)) of the motor (60). Thus, in order to obtain the estimated angular velocity value (ω _e2 (t)) of the motor 60 using both the current torque estimated value (τ _e2 (t)) and the measured torque value (τ _s2 (t)), A precise angular velocity estimate (ω _e2 (t)) can be obtained.

Further, the motor control device (30) mutually influences the load (40) connected to the motor (60) between the first element (1) and the first element (1). Modeled into a second element (2, 50) that exerts and interlocks with the motor (60), and based on the current torque estimate (.tau. _E2 (t)), the angular velocity of the first element (1) A load model unit (10) for outputting the estimated value (ω _e1 (t)) is further characterized. Thereby, the angular velocity estimated value (ω _e1 (t)) of the first element (1) can be obtained based on the model of the load.

Further, the future torque calculation unit (12) comprises an estimated angular velocity value (ω _e2 (t)) of the motor (60) and an estimated angular velocity value (ω _e1 (t)) of the first element (1). , And outputs the estimated future torque value (τ _e2 (t + Δt)). Thereby, the future torque estimated value (τ _e2 (t + Δt)) can be obtained based on the model of the load.

When the motor control device (30) receives the estimated angular velocity value (ω _e2 (t)) of the motor (60) and the measured angular velocity value (ω _s2 (t)) of the motor (60) A rotation compensation controller (A) that outputs a torque correction value (Δτ (t)) for asymptotically _{approaching the} estimated angular velocity value (ω _e2 (t)) of the motor (60) to the measured angular velocity value (ω _s2 (t)) The torque command value output unit (20) further includes an addition result of the future torque estimated value (τ _e2 (t + Δt)) and the torque correction value (Δτ (t)) as the torque command value. It is characterized in that it is outputted as (τ _v (t)). Thereby, the error of the angular velocity estimated value (ω _e2 (t)) of the motor (60) can be reduced.

The future torque estimated value (τ _e2 (t + Δt)) is a value corresponding to the torque to be realized by the motor in the future by a predetermined time difference (Δt) than the present, and the future torque calculator (12 ), the first element (1) of the angular velocity estimate (omega _e1 (t)) and the (angular velocity estimate of 60) (omega _e2 (t) motor) and the current angular difference estimates a difference ( [Delta] [omega _e (t)) and outputs the subtractor (70), the current difference in angular velocity estimate ([Delta] [omega _e (t) first by multiplying a predetermined value corresponding to the time difference (Delta] t) relative to) the (P) , A differentiator (72) for differentiating the output signal of the first proportional unit (71), the current angular velocity difference estimated value (.DELTA..omega. _E (t)) and the differentiator (72) And an output signal of the first adder (73). A spring constant (K) between the first element (1) and the second element (2, 50) is multiplied by the divider (81) and the output signal of the integrator (81) Between the first element (1) and the second element (2, 50) with respect to the output signal of the second adder (73) and the output signal of the first adder (73). A third proportional unit (84) for multiplying the viscous damping coefficient (C), an output signal of the second proportional unit (82) and an output signal of the third proportional unit (84) are added and added And a second adder (83) for outputting the result as the future torque estimated value (τ _e2 (t + Δt)). Thereby, the future torque estimated value (τ _e2 (t + Δt)) can be acquired with a simple configuration.

[Modification]
The present invention is not limited to the embodiments described above, and various modifications are possible. The embodiments described above are illustrated to facilitate understanding of the present invention, and are not necessarily limited to those having all the described configurations. Further, part of the configuration of one embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of one embodiment. In addition, it is possible to delete part of the configuration of each embodiment or to add / replace other configuration. Possible modifications to the above embodiment are, for example, as follows.

(1) In the above embodiment, an induction motor is applied as the absorption motor 60, but the absorption motor 60 may be another motor such as a synchronous motor or a direct current motor.
(2) dead time T _eb and time constant set in the emulation unit 14 T _ec is normally a dead time occurring T _b and the time constant T _c in the inverter 62 and the absorption motor 60, it is desirable to match . However, even if they do not match, it is possible to achieve better results than the comparative example.

(3) The hardware of the control device 30 in the above embodiment can be realized by a general computer, so programs etc. for realizing the algorithm shown in FIGS. 3 and 4 are stored in a storage medium or via a transmission line. You may distribute it.
(4) The algorithm shown in FIG. 3 and FIG. 4 is realized by software-like processing using a program in the above embodiment, but a part or all of the algorithm is realized by ASIC (Application Specific Integrated Circuit; application specific IC) Alternatively, the processing may be replaced by hardware processing using an FPGA (field-programmable gate array) or the like.

(5) In the above embodiment, the future torque calculator 12 obtains the future torque estimated value τ _e2 (t + Δt) by the algorithm shown in FIG. 4, but the method for determining the future torque estimated value τ _e2 (t + Δt) is It is not necessarily limited to this. For example, the angular acceleration of the element 2 may be multiplied by the time difference Δt, and the current angular velocity estimated value ω _e2 (t) may be added to the multiplication result to obtain the angular velocity difference estimated value Δω _e (t + Δt). Further, the estimated angular velocity values ω _e1 (t) and ω _e2 (t) are respectively multiplied by the time difference Δt, and these multiplication results are added to the current angles of the elements 1 and 2 to obtain an element 1, 1 in the future by the time difference Δt. An angle of 2 may be calculated, and the future torque estimated value τ _e2 (t + Δt) may be determined based on the difference in angle between the elements 1 and 2 in the future.

(6) In the above embodiment, the characteristic test of the completed automobile 40 is performed. However, the present invention is not limited to the test system in which the absorbing motor is connected to the parts of the automobile 40 and the characteristic test of each part is performed. The invention may be applied. For example, when only the engine 41 is completed, the absorption motor can be connected to the engine 41 and the engine 41 can be tested alone. When only the transmission 42 is completed, a motor simulating the engine 41 and a pair of absorption motors can be connected to the transmission 42 to test the transmission 42 alone. Since precise torque control is required for any of these absorption motors and motors that simulate the engine 41, it is preferable to apply the present invention to perform these torque controls.

(7) Moreover, the present invention can be applied not only to the test of the automobile or its parts but also to various applications where precise torque control is required. For example, the present invention may be applied to industrial robots, machine tools, construction machines and the like.

1 element (first element)
Two elements (second element)
10 Vehicle model section (load model section)
12 Future torque calculation unit 14 Emulation unit 16 Wheel speed calculation unit 18 Rotation compensation controller 20 Adder (torque command value output unit)
30 Controller (Motor controller, computer)
37 Output interface 38 Input interface 40 Automobile (load)
50, 50 FL, 50 FR, 50 RL, 50 RR tire type bearing (second element)
52, 52 FL, 52 FR, 52 RL, 52 RR Sensor unit 60, 60 FL, 60 FR, 60 RL, 60 RR Absorbing motor (motor)
62, 62 FL, 62 FR, 62 RL, 62 RR Inverter 70 Subtractor 71 Proportioner (First Proportioner)
72 Differentiator 73 Adder (First Adder)
80 spring calculator 81 integrator 82 proportional (second proportional)
83 Adder (2nd Adder)
84 Proportioner (Third Proportioner)
C Viscosity damping coefficient K Spring constant P Predetermined value Δt Time difference Δτ (t) Torque correction value Δω _e (t) Angular velocity difference estimated value (current angular velocity difference estimated value)
τ _v (t) Torque command value τ _e2 (t) Present torque estimated value τ _e2 (t + Δt) Future torque estimated value τ _s2 (t) Torque measured value ω _e1 (t) Angular velocity estimated value ω _e2 (1) of the first element t) Angular velocity estimated value ω _s2 (t) of second element

Claims (8)

A motor control device that supplies the torque command value to an inverter that applies a drive voltage to a motor based on a supplied torque command value,
A future torque calculation unit that calculates a future torque estimated value corresponding to a torque estimated to be realized in the future by the motor;
A torque command value output unit that outputs the torque command value based on the future torque estimated value ;
A load model unit modeling a load connected to the motor into a model including a first element, and outputting an estimated angular velocity value of the first element;
Have
The motor control apparatus according to claim 1, wherein the future torque calculation unit outputs the future torque estimated value based on the angular velocity estimated value of the motor and the angular velocity estimated value of the first element . The motor control device according to claim 1, further comprising: an emulation unit that emulates the operation of the motor and the inverter and outputs an estimated current torque value of the motor based on the estimated future torque value. . The speed calculating unit according to claim 2, further comprising: a speed calculation unit that outputs an estimated angular velocity value of the motor based on the current torque estimated value and a torque measurement value that is a result of measuring the torque of the motor. Motor controller. The load model unit models the load into the first element and a second element that influences each other between the first element and interlocks with the motor, and estimates the current torque. Outputting the estimated angular velocity value of the first element based on the value
The motor control device according to claim 3, characterized in that. And a rotation compensation controller for outputting a torque correction value for making the estimated angular velocity value of the motor asymptotically approach the measured angular velocity value when the estimated angular velocity value of the motor and the measured angular velocity value of the motor are input. ,
The motor control device according to claim 4 , wherein the torque command value output unit outputs an addition result of the future torque estimated value and the torque correction value as the torque command value. The future torque estimated value is a value corresponding to the torque to be realized by the motor in the future by a predetermined time difference from the present,
The future torque calculation unit
A subtractor which outputs a current angular velocity difference estimated value which is a difference between the first estimated angular velocity value of the first element and the estimated angular velocity value of the motor;
A first proportional unit for multiplying the current angular velocity difference estimated value by a predetermined value corresponding to the time difference;
A differentiator that differentiates the output signal of the first proportional element;
A first adder for adding the current angular velocity difference estimated value and the output signal of the differentiator;
An integrator that integrates the output signal of the first adder;
A second proportional unit that multiplies the spring constant between the first element and the second element with the output signal of the integrator;
A third proportional unit which multiplies the viscous damping coefficient between the first element and the second element by the output signal of the first adder;
A second adder for adding the output signal of the second proportional unit and the output signal of the third proportional unit and outputting the addition result as the estimated future torque value. The motor control device according to 5 . A computer that supplies the torque command value to an inverter that applies a drive voltage to the motor based on the supplied torque command value;
A future torque calculation unit that calculates a future torque estimated value corresponding to a torque estimated to be realized in the future by the motor,
A torque command value output unit that outputs the torque command value based on the future torque estimated value;
A load model unit modeling a load connected to the motor into a model including a first element, and outputting an estimated angular velocity value of the first element;
To act as
The future torque calculation unit outputs the future torque estimated value based on the angular velocity estimated value of the motor and the angular velocity estimated value of the first element.
A program characterized by Furthermore, the computer
An emulation unit that emulates the operation of the motor and the inverter, and outputs a current torque estimate of the motor based on the future torque estimate;
A speed calculator configured to output an estimated angular velocity value of the motor based on the current torque estimated value and a measured torque value that is a result of measuring the torque of the motor;
To act as
The load model unit, the load is modeled with the first element, the second element is linked to the motor with influence on each other between the first element, the current torque estimation based on the value, the program according to claim 7, characterized in that you output an angular velocity estimated value of the first element.

Images