JP7635670B2 - Rotating electric machine control method and rotating electric machine control system - Google Patents

Description

The present invention relates to a rotating electric machine control method and a rotating electric machine control system.

Patent document 1 describes a motor control device that corrects the rotor position detected by a resolver and uses it to control the motor. This motor control device corrects rotor position detection errors by software processing. More specifically, the motor control device of Patent document 1 predicts the detection value based on the average value of the time required for the detection value by the resolver to change by a predetermined value, and corrects the actual detection value based on this prediction. In addition, the corrected rotor position information is used for coordinate transformation that converts the three-phase motor currents (Iv, Iv, Iw) into d-axis and q-axis currents (Id, Iq), and for coordinate transformation that converts the d-axis and q-axis voltage control amounts (Vd, Vq) into three-phase voltage control amounts (Vu, Vv, Vw).

JP 2010-283925 A

One of the methods for synchronous control of rotating electrical machines is phase-referenced synchronous PWM (Pulse Width Modulation) control, which uses the rotor position detection signal as a carrier signal. However, the rotor position detection signal contains detection errors. For this reason, in phase-referenced synchronous PWM control, errors in the rotor position detection signal used as the carrier signal can cause non-negligible harmonics and ripples in the rotating electrical machine current.

In addition, when performing phase-reference synchronous PWM control, the position detection signal obtained from a position detector such as a resolver is input directly to the PWM signal generator without software processing, so the position detection signal used as the carrier signal is not corrected by software processing. Therefore, when the rotor position detection signal is used as the carrier signal for PWM control, even if the carrier signal contains rotor position detection errors, this cannot usually be corrected. As a result, phase-reference synchronous PWM control has the problem that it is not possible to reduce harmonics and ripples caused by rotor position detection errors, and the risk of overcurrent and overheating in the rotating electric machine cannot be completely eliminated.

The present invention aims to provide a rotating electric machine control method and rotating electric machine control system that can reduce current harmonics and ripples caused by rotor position detection errors when using a rotor position detection signal as a carrier signal for PWM control.

One aspect of the present invention is a method for controlling a rotating electric machine by a rotating electric machine control system including a first control device that calculates parameters for controlling the driving of the rotating electric machine in response to a drive request for the rotating electric machine, and a second control device that executes synchronous PWM control to generate a PWM signal corresponding to a drive frequency of the rotating electric machine using the parameters and a rotor position detection signal of the rotating electric machine. In this rotating electric machine control method, the second control device acquires a rotor position detection signal, and the second control device generates a carrier signal for the synchronous PWM control based on the position detection signal. Meanwhile, the first control device acquires a position-dependent error, which is an error in rotor position detection and is determined according to the rotor position. In addition, the first control device acquires a basic comparison value to be compared with the carrier signal, and the first control device calculates a corrected comparison value by correcting the basic comparison value using the position-dependent error. Then, the second control device generates a PWM signal by comparing the carrier signal with the corrected comparison value. Moreover, the calculation of the correction comparison value and the generation of the PWM signal using the correction comparison value are executed when the modulation rate of the PWM signal is greater than a predetermined threshold value.

The rotating electric machine control method and rotating electric machine control device of the present invention can reduce current harmonics and ripples caused by rotor position detection errors when using a rotor position detection signal as a carrier signal for PWM control.

FIG. 1 is a block diagram showing the configuration of a rotating electrical machine control system. FIG. 2 is a block diagram showing the configuration of the synchronous PWM control unit. FIG. 3 is a block diagram showing the configuration of the modulation switching determination unit. FIG. 4 is a flowchart showing the operation of the rotating electrical machine control system. FIG. 5 is a timing chart showing the changes over time of the carrier signal and the PWM signal in the basic asynchronous PWM control and synchronous PWM control. FIG. 6 is an explanatory diagram showing the influence of the position-dependent error contained in the position detection signal and the effect of the correction comparison value Th[n]. FIG. 7 is a block diagram showing the configuration of a rotating electrical machine control system according to the second embodiment. FIG. 8 is a block diagram showing the configuration of the synchronous PWM control unit 15 in the second embodiment. FIG. 9 is a timing chart showing changes over time of the carrier signal and the synchronous PWM signal in the second embodiment. FIG. 10 is an explanatory diagram showing the operation of the rotating electrical machine control system according to the second embodiment. FIG. 11 is a block diagram showing the configuration of a rotating electrical machine control system according to the third embodiment. FIG. 12 is a block diagram showing the configuration of a synchronous PWM control unit in the third embodiment. FIG. 13 is a block diagram showing a configuration for an error correction unit of the fourth embodiment to calculate errors relating to position detection. FIG. 14 is a graph illustrating the relationship between the rotor position index array and the position dependent error array. FIG. 15 is a graph showing the relationship between the true rotor position and the position-dependent error obtained by directly referring to the rotor position index array and the position-dependent error array based on the position detection signal. FIG. 16 is a graph showing the relationship between the difference array and the error array.

The following describes an embodiment of the present invention with reference to the drawings.

[First embodiment]
Fig. 1 is a block diagram showing a configuration of a rotating electric machine control system 100. As shown in Fig. 1, the rotating electric machine control system 100 is a system for controlling a rotating electric machine 101, and includes a first control device 102, a second control device 103, an inverter 104, and the like.

The rotating electric machine 101, which is the object of control of the rotating electric machine control system 100, is a motor (electric motor) and/or a generator (electric generator). In either case, the rotating electric machine 101 is composed of a stator and a rotor, etc., not shown. The rotating electric machine control system 100 is mounted on a vehicle such as a hybrid vehicle or an electric vehicle. In this embodiment, the rotating electric machine control system 100 is mounted on an electric vehicle, not shown, and the rotating electric machine 101 is a drive motor for driving the electric vehicle. The rotating electric machine 101 is, for example, an IPM (Interior Permanent Magnet) type three-phase AC motor.

The first control device 102 is a so-called vehicle controller, and is composed of one or more computers that calculate parameters for controlling the driving of the rotating electric machine 101 in response to a drive request for the rotating electric machine 101. The computer that constitutes the first control device 102 is composed of a CPU (Central Processing Unit), a memory, etc. The first control device 102 is also programmed to calculate parameters for controlling the driving of the rotating electric machine 101 at a predetermined control period τ in response to a drive request for the rotating electric machine 101. In other words, the first control device 102 is a calculation device that calculates parameters used for PWM control of the rotating electric machine 101 according to software consisting of one or more programs. The drive request for the rotating electric machine 101 is a request regarding the driving state of the rotating electric machine 101, and is a command that determines the torque of the rotating electric machine 101, or a collection of these. The predetermined control period τ in the rotating electric machine control system 100 is, for example, 100 μs.

The first control device 102 calculates parameters for controlling the rotating electric machine 101 by PWM control based on a torque command value T ^* , which is a target value of the torque to be output by the rotating electric machine 101, and a rotor position detection signal θ of the rotating electric machine 101. The torque command value T ^* is calculated by a higher-level vehicle controller or control block (not shown) according to the accelerator opening degree of the electric vehicle, etc. The rotor position detection signal θ is an electrical angle indicating the rotational position of the rotor, and is acquired by the second control device 103 and input to the first control device 102 as appropriate.

The PWM control parameters calculated by the first control device 102 include, for example, the three-phase duty command values Duty_u, Duty_v, and Duty_w in asynchronous PWM control, the correction comparison value Th[n] in synchronous PWM control, and the modulation mode signal Sm for switching between asynchronous PWM control and synchronous PWM control. Details of these various parameters will be described later. The various PWM control parameters calculated by the first control device 102 are input to the second control device 103.

Note that asynchronous PWM control is PWM control that uses a PWM signal (hereinafter referred to as an asynchronous PWM signal) with a predetermined frequency that is independent of the drive frequency of the rotating electric machine 101. In this embodiment, a triangular wave with a frequency of 1/200 μs is used as the carrier signal for asynchronous PWM control.

The synchronous PWM control is a PWM control that uses a PWM signal (hereinafter, referred to as a synchronous PWM signal) according to the drive frequency of the rotating electric machine 101. In the synchronous PWM control, for example, a synchronous PWM signal is generated so as to be proportional to the drive frequency of the rotating electric machine 101. The drive frequency of the rotating electric machine 101 is expressed, for example, by the rotor position detection signal θ or other parameters that have a certain correlation with the rotor position detection signal θ. Therefore, the rotor position detection signal θ, the sum (θ+α ^* ) of the rotor position detection signal θ and the voltage phase α ^* , or a signal obtained by folding back these at a predetermined period, is used as a carrier signal for the synchronous PWM control. In this embodiment, the sum of the rotor position detection signal θ and the voltage phase α ^* is used as the carrier signal for the synchronous PWM control.

The second control device 103 executes PWM control of the rotating electric machine 101 by inputting drive signals D _uu * to D _wl * for driving switching elements to the inverter 104 according to the PWM control parameters calculated by the first control device 102. Specifically, ^the second control device 103 executes asynchronous PWM control by inputting drive signals D _uu ^* to D _wl ^* generated based on duty command values Duty_u, Duty_v, and Duty_w ^, which are parameters for asynchronous PWM control, to the inverter 104. The second control device 103 also executes synchronous PWM control by inputting drive signals D _uu ^* to D _wl ^* generated by comparing a correction comparison value Th[n], which is a parameter for synchronous PWM control, with a carrier signal for synchronous PWM control, to the inverter 104.

The second control device 103 also performs signal processing and other operations required to perform asynchronous PWM control and synchronous PWM control. For example, the carrier signal used for these PWM controls is generated by the second control device 103.

The second control device 103 is, for example, a controller configured with a microcomputer or other signal processing circuit device that performs signal processing specialized for executing PWM control. The second control device 103 is, for example, configured separately from the first control device 102. In the rotating electric machine control system 100, the second control device 103 is a sub-control device that performs specific signal processing according to parameters specified by the first control device 102, which is the main control device. In addition, when the rotating electric machine control system 100 is viewed as a whole, the first control device 102 is a general-purpose calculation area that performs various calculations according to software, and the second control device 103 is a circuit for PWM control (so-called PWM timer, etc.) that is connected to the first control device 102. Therefore, it can be said that the first control device 102 is the CPU area of the rotating electric machine control system 100, and the second control device 103 is the peripheral area (peripheral device) of the rotating electric machine control system 100.

In this way, because the second control device 103 is a peripheral area, the calculations of the first control device 102, which is a CPU area, do not affect the signal processing within the second control device 103. For example, because the carrier signal used for PWM control is generated within the second control device 103 and used within the second control device 103, the first control device 102 cannot modulate or correct the carrier signal used for PWM control.

The inverter 104 drives the built-in switching elements according to the drive signals D _uu ^* to D _wl ^* output by the second control device 103. As a result, the inverter 104 converts the DC voltage of the battery 105 (hereinafter referred to as the battery voltage V _dc ) into three-phase AC voltages v _u , v _v , and v _w and inputs them to the rotating electric machine 101. As a result, the rotating electric machine 101 is driven to output a torque that matches the torque command value T ^* . A DC voltage sensor 106 is provided at the input end of the battery 105. Therefore, the battery voltage V _dc is detected by the DC voltage sensor 106 at any timing.

<Specific Configuration of First Control Device>
The first control device 102 includes a torque control unit 11 , a coordinate conversion unit 12 , an asynchronous PWM control unit 13 , a vector conversion unit 14 , a synchronous PWM control unit 15 , and a modulation switching determination unit 16 .

The torque control unit 11 calculates a d-axis voltage command value vd _* and a q-axis voltage command value vq _* based on the torque command value _T ^* , the rotation speed N of the rotating electric machine 101, the battery voltage _Vdc ^, and the d-axis current detection value id and the q-axis current detection value iq ^of the rotating electric machine 101. The d-axis voltage command value _vd ^* and the q-axis voltage _command value _vq ^* are calculated by a current vector control (CVC) calculation or a voltage phase control (VPC) calculation. The calculated d-axis voltage command value _vd ^* and q-axis voltage command value _vq ^* are input to the coordinate conversion unit 12 and the vector conversion unit 14.

The coordinate converter 12 converts the d-axis voltage command value _vd ^* and the q-axis voltage command value _vq ^* into three-phase voltage command values _vu ^* , _vv ^* , and _vw ^* based on a read-ahead position _θL , which will be described later. This conversion is performed according to the following equation (1).

The asynchronous PWM control unit 13 calculates duty command values Duty_u, Duty_v, and Duty_w, which are parameters for asynchronous PWM control, based on the three-phase voltage command values _vu ^* , _vv ^* , and _vw ^* and the battery voltage _Vdc . The duty command value Duty_u for the U phase is calculated according to the following formula (2). The same is true for the duty command values Duty_v and Duty_w for the V and W phases. The duty command values Duty_u, Duty_v, and Duty_w are expressed in units of percent (%). The duty command values Duty_u, Duty_v, and Duty_w for each phase calculated by the asynchronous PWM control unit 13 are input to the second control device 103.

The vector conversion unit 14 calculates the voltage norm V _a ^* and the voltage phase α ^* , which are parameters of the synchronous PWM control, based on the d-axis voltage command value v _d ^* and ^{the q-axis voltage command value v q} _* . The voltage norm V _a ^* and the voltage phase α ^* are calculated according to the following formula (3). The voltage norm V _a ^* represents the norm of the voltage to be applied to the rotating electric machine 101 in response to the drive request. Similarly, the voltage phase α ^* represents the phase of the voltage to be applied to the rotating electric machine 101 in response to the drive request.

The synchronous PWM control unit 15 calculates a correction comparison value Th[n], which is a parameter of the synchronous PWM control, based on the voltage norm V _a ^* , the position-dependent error Δθ _L , and the battery voltage V _dc. The position-dependent error Δθ _L is an error in detecting the position of the rotor, and is determined according to the position of the rotor. The position-dependent error Δθ _L is acquired by an error generating unit 20, which will be described later. The correction comparison value Th[n] is one or more threshold values that are compared with the carrier signal in order to set the on/off timing of the synchronous PWM signal in the synchronous PWM control. That is, in the synchronous PWM control, the on/off of the PWM signal is set by a compare match between the correction comparison value Th[n] and the carrier signal. The correction comparison value Th[n] calculated by the synchronous PWM control unit 15 is input to the second control device 103. Note that "n" is an integer, and is an even number in the normal case where there is no folding back of the carrier signal.

Figure 2 is a block diagram showing the configuration of the synchronous PWM control unit 15. As shown in Figure 2, more specifically, the synchronous PWM control unit 15 includes a modulation rate calculation unit 31, a basic comparison value table 32, and a comparison value correction unit 33.

The modulation factor calculation unit 31 calculates the modulation factor M of the synchronous PWM signal based on the voltage norm V _a ^* and the battery voltage V _dc . That is, the modulation factor M is set according to a drive request for the rotating electric machine 101. Specifically, the modulation factor M is calculated according to the following formula (4).

The basic comparison value table 32 is a table that stores a basic comparison value Th' [n] in association with each modulation factor M. Therefore, the synchronous PWM control unit 15 functions as a comparison value acquisition unit that acquires the basic comparison value Th' [n] corresponding to the modulation factor M by referring to the basic comparison value table 32. The basic comparison value Th' [n] is determined in advance for each modulation factor M by experiments or numerical calculations using SHE (specific harmonic elimination method) so as to suppress the harmonic current and ripple of the rotating electric machine 101. However, the basic comparison value Th' [n] is a basic comparison value that is based on the assumption that the rotor position detection signal θ is compared with a carrier signal in an ideal state in which no error is substantially included. In other words, the basic comparison value Th' [n] determines the on/off phase of the synchronous PWM signal with respect to the carrier signal in an ideal state.

In this embodiment, the basic comparison value table 32 stores 10 basic comparison values Th'[1] to Th'[10] for each modulation factor M. Therefore, the synchronous PWM control of this embodiment can generate a synchronous PWM signal that includes up to five pulses per period of the electrical angle (position detection signal θ), which is the carrier signal.

The comparison value correction unit 33 calculates a corrected comparison value Th[n] by correcting the basic comparison value Th'[n] acquired based on the modulation factor M according to the position-dependent error Δθ _L. In this embodiment, the comparison value correction unit 33 is composed of an adder 34 and a mask processing unit 35. The adder 34 adds the position-dependent error Δθ _L to the basic comparison value Th'[n]. The mask processing unit 35 receives the basic comparison value Th'[n] to which the position-dependent error Δθ _L has been added as an input (in), and performs a mask process to assign the basic comparison value Th'[n] to which the position-dependent error Δθ _L has been added to a range of 0 to 2π using a modulus operation (MOD) with 2π as a divisor. As a result, the comparison value correction unit 33 outputs the corrected comparison value Th[n] corrected based on the position-dependent error Δθ _L.

Although the comparison value corrector 33 in this embodiment calculates the corrected comparison value Th[n] by adding the position-dependent error Δθ _L , the correction comparison value Th[ _n ] can be calculated by other calculation methods depending on the setting of the position-dependent error Δθ L. For example, the comparison value corrector 33 can calculate the corrected comparison value Th[n] by subtracting, multiplying, or dividing the position-dependent error Δθ _L from the basic comparison value Th'[n], or by calculating according to other functions, etc.

In addition, in this embodiment, the basic comparison value Th'[n] and the corrected comparison value Th[n] are common to each of the UVW phases.

The modulation switching determination unit 16 (see FIG. 1) determines whether the rotating electric machine 101 should be driven by asynchronous PWM control or synchronous PWM control, based on the voltage norm V _a ^* and the battery voltage V _dc . The determination result of the modulation switching determination unit 16 is input to the second control device 103 as a modulation mode signal Sm. The modulation mode signal Sm is a command that specifies the modulation method of the PWM signal to be either asynchronous PWM control or synchronous PWM control.

Figure 3 is a block diagram showing the configuration of the modulation switching determination unit 16. As shown in Figure 3, the modulation switching determination unit 16 includes a modulation rate calculation unit 36 and a modulation method selection unit 37.

The modulation factor calculation unit 36 is configured similarly to the modulation factor calculation unit 31 included in the synchronous PWM control unit 15. That is, the modulation factor calculation unit 36 calculates the modulation factor M of the synchronous PWM signal based on the voltage norm V _a ^* and the battery voltage V _dc in accordance with the above-mentioned equation (4).

The modulation method selection unit 37 selects the modulation method of the PWM signal from either asynchronous PWM control or synchronous PWM control by comparing the modulation rate M with a predetermined threshold. In this embodiment, the modulation method selection unit 37 selects the modulation method by providing hysteresis to the switching between asynchronous PWM control and synchronous PWM control, as shown in Table 1 below. Specifically, when performing asynchronous PWM control, if the modulation rate M exceeds 1.0 and control in the so-called overmodulation region or rectangular wave region is required, the modulation method selection unit 37 selects synchronous PWM control. On the other hand, when performing synchronous PWM control, if the modulation rate M falls below 0.9 and control in the so-called normal modulation region is sufficient, the modulation method selection unit 37 selects asynchronous PWM control. Therefore, when switching from asynchronous PWM control to synchronous PWM control, "1.0" is the predetermined threshold value to be compared with the modulation rate M. Also, when switching from synchronous PWM control to asynchronous PWM control, "0.9" is the predetermined threshold value to be compared with the modulation rate M. That is, the rotating electrical machine control system 100 has an asynchronous PWM control mode in which asynchronous PWM control is performed, and a synchronous PWM control mode in which synchronous PWM control is performed, and the asynchronous PWM control mode and the synchronous PWM control mode are switched depending on whether the modulation factor M exceeds a predetermined threshold. In this embodiment, the synchronous PWM control mode is performed when the modulation factor M is greater than the predetermined threshold.

In addition to the above, the first control device 102 includes an error correction unit 17, a coordinate conversion unit 18, a look-ahead compensation unit 19, an error generation unit 20, and a rotation speed calculation unit 21 (see FIG. 1).

The error correction unit 17 acquires the rotor position detection signal θ from the second control device 103. Then, the error correction unit 17 calculates a corrected position detection signal θc by correcting the error contained in the position detection signal _θ .

The errors contained in the position detection signal θ include a first-order error, a second-order error, and a structure-related order error. The first-order error is an error that varies in the first order with respect to the electrical angle (position detection signal θ). The second-order error is an error that varies in the second order with respect to the electrical angle (position detection signal θ). The structure-related order error is an error caused by the structure of the rotating electric machine 101 and the structure of the resolver 22 that detects the position of the rotor of the rotating electric machine 101. More specifically, the structure-related order error is an error that varies in the order according to the number of pole pairs of the rotating electric machine 101 and the number of teeth of the stator of the resolver 22 (hereinafter referred to as the resolver stator) with respect to the electrical angle (position detection signal θ). All of these errors contained in the position detection signal θ vary according to the position (electrical angle) of the rotor of the rotating electric machine 101. Therefore, some or all of these errors are, in a broad sense, position-dependent errors.

The position-dependent error Δθ calculated by the error correction unit 17 and the position-dependent error _ΔθL used by the synchronous PWM control unit 15 and the like are common in that they both vary according to the position of the rotor of the rotating electric machine 101, but they are different. Specifically, the position-dependent error Δθ calculated by the error correction unit 17 is calculated based on the position detection signal θ. On the other hand, the position-dependent error _ΔθL used by the synchronous PWM control unit 15 and the like is calculated based on a read-ahead position _θL , as described later. Therefore, hereinafter, the position-dependent error Δθ calculated by the error correction unit 17 is simply referred to as the "error Δθ" to distinguish it from the position-dependent error _ΔθL used by the synchronous PWM control unit 15 and the like.

In this embodiment, the error correction unit 17 calculates the error Δθ according to the following equation (5).Then, the error correction unit 17 calculates the corrected position detection signal θc by subtracting the error Δθ from the position detection signal _θ as shown in the following equation (6).

The first term on the right side of equation (5) is a first-order error. That is, the first-order error can be expressed as a sine wave with an amplitude A ₁ and a reference phase δ ₁ with respect to the electrical angle (position detection signal θ). However, the first-order error may be expressed in a form in which the phase is offset based on the rotation speed N and the control period τ. The first-order error may also be expressed as a cosine wave.

The second term on the right side of equation (5) is a second-order error. That is, the second-order error can be expressed as a sine wave with an amplitude _A2 and a reference phase _δ2 with respect to the electrical angle (position detection signal θ). The second-order error may be expressed in a form in which the phase is offset based on the rotation speed N and the control period τ, or may be expressed as a cosine wave.

The third term on the right side of the formula (5) is the structural order error. If the number of pole pairs of the rotating electric machine 101 is "p" and the number of teeth of the resolver stator is "m", the structural order error appears in the order of m/p with respect to the electrical angle (position detection signal θ). In this embodiment, the number of pole pairs of the rotating electric machine 101 is 4, and the number of teeth of the resolver stator is 10. Therefore, the structural order error in this embodiment varies in the order of 2.5 with respect to the electrical angle (position detection signal θ). Therefore, the structural order error can be expressed by a sine wave with an amplitude A _{of 2.5} and a reference phase δ of _2.5 with respect to the electrical angle (position detection signal θ). In addition, the structural order error may be expressed in a form in which the phase is offset based on the rotation speed N and the control period τ, or may be expressed by a cosine wave.

Therefore, as shown in equation (5), the error Δθ in this embodiment is obtained by a calculation that synthesizes a sine wave or a cosine wave based on the amplitudes A ₁ , A ₂ , A _2.5 and reference phases δ ₁ , δ ₂ , δ _2.5 of the first-order error, second-order error, and structure-related order error.

The amplitudes _A1 , _A2 , _A2.5 and reference phases _δ1 , _δ2 , _δ2.5 of the first-order error, second-order error, and structure-related order error can be obtained in advance at the time of manufacturing or shipping the rotating electric machine control system 100. For example, the amplitudes _A1 , _A2 , _A2.5 and reference phases _δ1 , _δ2 , _δ2.5 of each error can be obtained by performing fast Fourier transform processing (FFT processing) on the position detection signal θ for each rotation speed N of the rotating electric machine 101.

In this embodiment, as in the above formula (5), the error Δθ is the sum of the first-order error, the second-order error, and the structure-related order error, but the error Δθ may be any of the first-order error, the second-order error, or the structure-related order error. The error Δθ may also be calculated by a partial combination of the first-order error, the second-order error, or the structure-related order error. In addition, similar to the first-order and second-order errors, errors that vary at third or higher orders with respect to the electrical angle (position detection signal θ) may also be taken into account.

The coordinate conversion unit 18 converts the three-phase current detection values into a d-axis current detection value i _d and a q-axis current detection value i _q based on the corrected position detection signal θ _c . The current sensor 23 detects at least two-phase currents at the output terminal of the inverter 104. In this embodiment, the current sensor 23 detects a U-phase current and a V-phase current. Therefore, in this embodiment, the coordinate conversion unit 18 converts the U-phase current detection value i _u and the V-phase current detection value i _v into a d-axis current detection value i _d and a q-axis current detection value i _q according to the following equation (7). The d-axis current detection value i _d and the q-axis current detection value i _q are used by the torque control unit 11 as described above.

The look-ahead compensation unit 19 further compensates the corrected position detection signal θ _c based on the rotation speed N, taking into consideration the control period τ and the like, to calculate a compensated corrected position detection signal θ _c (hereinafter referred to as a look-ahead position θ _L ). In this embodiment, since the control period τ is 100 μs, the voltage based on the d-axis voltage command value v _d ^* and the q-axis voltage command value v _q ^* calculated by the torque control unit 11 is applied to the rotating electric machine 101 from about 100 μs to about 200 μs later. Therefore, the look-ahead compensation unit 19 estimates the position of the rotor 101 at 150 μs later, which is the intermediate value between them. Specifically, the look-ahead compensation unit 19 compensates the corrected position detection signal θ _c according to the following formula (8) to calculate the look-ahead position θ _L. In formula (8), "p" is the number of pole pairs of the rotating electric machine 101. The look-ahead position θ _L is used by the coordinate conversion unit 12 and the error generation unit 20 described above.

The error generating unit 20 executes a calculation to generate a position-dependent error Δθ _L corresponding to the read-ahead position θ _L. Specifically, the error generating unit 20 generates the position-dependent error Δθ _L according to the following equation (9). That is, the error generating unit 20 functions as a position-dependent error acquiring unit that acquires the position-dependent error Δθ _L.

Equation (9) is similar to the above-mentioned equation (5) except that it uses the look-ahead position θ _L instead of the position detection signal _θ and calculates a position-dependent error Δθ _L , which is an error corresponding to the look-ahead position θ _L. Therefore, the first term on the right side is a first-order error corresponding to the look-ahead position θ _L , and the second term on the right side is a second-order error corresponding to the look-ahead position θ _L. Similarly, the third term on the right side is a structure-attributable order error corresponding to the look-ahead position θ L. Therefore, each term in equation (9) may be expressed in a form in which the phase is offset based on the rotation speed N and the control period τ, or may be expressed as a cosine wave. In addition, the amplitudes A ₁ , A ₂ , and A _2.5 and the reference phases δ ₁ , δ ₂ , and δ _2.5 of the first-order error, second-order error, and structure-attributable order error are the same as those used in equation (5).

As described above, the position-dependent error Δθ _L generated by the error generating unit 20 is used in the comparison value correcting unit 33 of the synchronous PWM control unit 15 to correct the basic comparison value Th'[n] and calculate the corrected comparison value Th[n].

The rotation speed calculation unit 21 calculates the rotation speed N [rpm] of the rotating electric machine 101 based on the corrected position detection signal _θc . Specifically, the rotation speed calculation unit 21 converts the corrected position detection signal _θc into the rotation speed _N by calculating the amount of change per unit time of the corrected position detection signal θc according to the following equation (10). The rotation speed N is used by the torque control unit 11, the coordinate conversion unit 18, and the look-ahead compensation unit 19 as described above.

<Specific Configuration of Second Control Device>
The second control device 103 includes an asynchronous PWM signal generator 41, a synchronous PWM signal generator 42, a PWM output switch 43, and an ABZ counter 44 (see FIG. 1).

The asynchronous PWM signal generator 41 generates the asynchronous PWM signals Duua* _{to Dwla} ^* by a so-called triangular wave comparison method using the three-phase duty command values Duty_u, Duty_v, and Duty_w input from the asynchronous PWM control _unit ¹³ of the first control device 102. In the triangular wave comparison method, a triangular wave having a predetermined frequency independent of the drive frequency (position detection signal θ) of the rotating electric machine 101 is used as a carrier signal. More specifically, the asynchronous PWM signal generator 41 generates the asynchronous PWM signals _Duua ^* to _Dwla ^* by comparing and matching the three-phase duty command values Duty_u, Duty_v, and Duty_w with the triangular wave carrier signal _Ca.

In asynchronous PWM control, the arrangement interval of pulses in the asynchronous PWM signal is limited to the control period τ. The first subscripts "u", "v", and "w" of the asynchronous PWM signals D _uua ^* to D _wla ^* represent the respective phases UVW. The second subscripts "u" and "l" represent the upper arm element ("u") or lower arm element ("l") of the inverter 104. The third subscript "a" represents an asynchronous PWM signal.

The synchronous PWM signal generator 42 generates a carrier signal used in the synchronous PWM control based on the drive frequency (position detection signal θ) of the rotating electric machine 101. That is, the synchronous PWM signal generator 42 functions as a carrier signal generator that generates a carrier signal in the synchronous PWM control based on the position detection signal θ, etc. The synchronous PWM signal generator 42 also generates the synchronous PWM signals D _uus ^* to D _wls * by comparing and matching the corrected comparison value Th[n] input from the synchronous PWM control unit 15 of the first control device ¹⁰² with the generated carrier signal.

In this embodiment, the synchronous PWM signal generator 42 uses the voltage phase α ^* in addition to the position detection signal θ, and generates carrier signals C _us , C _vs , and C _ws for the UVW phases according to the following equation (11).

In the synchronous PWM control, the arrangement interval of the pulses in the synchronous PWM signal is almost independent of the control period τ and can be adjusted substantially arbitrarily. Therefore, when the rotating electric machine 101 is driven in the overmodulation region and the rectangular wave region where the number of pulses is limited, the synchronous PWM control has the advantage that it is easier to reduce harmonics and ripples of the current compared to the asynchronous PWM control. In addition, the first subscripts "u", "v", and "w" of the synchronous PWM signals D _uus ^* to D _wls ^* represent each phase of UVW. In addition, the second subscripts "u" and "l" represent the upper arm element ("u") or the lower arm element ("l") of the inverter 104. The third subscript "s" represents a synchronous PWM signal.

The PWM output switch 43 selects either the asynchronous PWM signals D _uua ^* to D _wla ^* or the synchronous PWM signals D _uus ^* to D _wls ^* in accordance with the modulation mode signal Sm input from the modulation switching determination unit 16. Then, the PWM output switch 43 outputs the selected PWM signal to the inverter 104 as the driving signals D _uu ^* to D _wl ^* for the switching elements.

The ABZ counter 44 samples the up counter pulse A, down counter pulse B, and origin signal pulse Z input from the resolver digital converter integrated circuit (RDIC) 24 every control period τ. In this way, the ABZ counter 44 measures the rotor's electrical angle, i.e., the position detection signal θ. Therefore, the ABZ counter 44 is a position detection signal acquirer that acquires the rotor's position detection signal θ. The position detection signal θ is directly input to the synchronous PWM signal generator 42 in the second control device 103 to generate a carrier signal for synchronous PWM control. The position detection signal θ is also input to the error correction unit 17 of the first control device 102.

The resolver digital conversion circuit 24 outputs an excitation signal to the resolver 22, and receives a modulation signal from the resolver 22. The resolver digital conversion circuit 24 outputs an up counter pulse A, a down counter pulse B, and an origin signal pulse Z to the ABZ counter 44 based on the excitation signal and the modulation signal. In addition, in the rotating electric machine control system 100, an encoder other than the resolver 22 can be used to detect the rotor position. When an encoder other than the resolver 22 is used, the specific configuration for measuring the position detection signal θ can be changed as appropriate depending on the encoder used.

<Action>
The operation of the rotating electrical machine control system 100 configured as above will now be described.

Fig. 4 is a flowchart showing the operation of the rotating electric machine control system 100. As shown in Fig. 4, in step S101, the ABZ counter 44 of the second control device 103 acquires a position detection signal θ indicating the position of the rotor of the rotating electric machine 101. At this time, when the rotating electric machine 101 is controlled by synchronous PWM control, the position detection signal θ is directly input from the ABZ counter 44 to the synchronous PWM signal generator 42 in the second control device 103, and carrier signals C _us , C _vs , and C _ws for synchronous PWM control are generated.

In step S102, the current detection values _iu and _iv are acquired. Then, in step S103, the error correction unit 17 of the first control unit 102 calculates a corrected position detection signal θc by correcting the position detection signal θ based on the error _Δθ . In step S104, the rotation speed calculation unit 21 of the first control unit 102 calculates the rotation speed N based on the corrected position detection signal _θc . In step S105, the coordinate conversion unit 18 of the first control unit 102 converts the current detection values _iu and _iv into a d-axis current detection value _id and a q-axis current detection value _iq based on the corrected position detection signal _θc . In step S106, the look-ahead compensation unit 19 of the first control unit 102 further compensates the corrected position detection signal _θc in consideration of the control period τ to calculate a look-ahead position _θL .

Then, in step S107, the torque control unit 11 of the first control device 102 executes a torque control calculation based on the torque command value T ^* , the rotation speed N of the rotating electric machine 101, the battery voltage _Vdc , and the _d -axis current detection value id and the q-axis current detection value _iq of the rotating electric machine 101. This calculates a d-axis voltage command value _vd ^* and a q-axis voltage command value _vq ^* . In addition, in step S108, the modulation switching determination unit 16 of the first control device 102 determines whether the rotating electric machine 101 is to be controlled by asynchronous PWM control or synchronous PWM control, and outputs a modulation mode signal Sm representing the determination result to the PWM output switch 43 of the second control device 103.

In order to control the rotating electric machine 101 by the synchronous PWM control (step S109: YES), in step S110, the vector conversion unit 14 of the first control device 102 calculates a voltage norm V _a ^* and a voltage phase α ^* . Furthermore, the synchronous PWM control unit 15 of the first control device 102 calculates a basic comparison value Th'[n] in step S111, and calculates a position-dependent error Δθ _L in step S112. Furthermore, in step S113, the synchronous PWM control unit 15 of the first control device 102 corrects the basic comparison value Th'[n] by using the position-dependent error Δθ _L through the comparison value correction unit 33. This allows the correction comparison value Th[n], which is a parameter used in the synchronous PWM control of this embodiment, to be calculated.

In addition, in order to control the rotating electric machine 101 by asynchronous PWM control (step S109: NO), in step S115, the coordinate conversion unit 12 of the first control device 102 converts the d-axis voltage command value _vd ^* and the _q -axis voltage command value vq ^* into three-phase voltage command values _vu ^* , _vv ^* , _vw ^* . Then, in step S116, the asynchronous PWM control unit 13 of the first control device 102 calculates duty command values Duty_u, Duty_v, and Duty_w, which are parameters for asynchronous PWM control.

Therefore, when the modulation mode signal Sm is set to execute synchronous PWM control, the second control unit 103 executes synchronous PWM control by comparing and matching the corrected comparison value Th[n] with the carrier signals _Cus , _Cvs , and _Cws in step S114. When the modulation mode signal Sm is set to execute asynchronous PWM control, the second control unit 103 executes asynchronous PWM control by comparing and matching the duty command values Duty_u, Duty_v, and Duty_w with the triangular wave carrier signal _Ca in step S117.

FIG. 5 is a timing chart showing the change over time of the basic carrier signal and PWM signal of the asynchronous PWM control and the synchronous PWM control. FIG. 5(A) shows the amplitude of the triangular wave carrier signal C _a , the duty command value Duty_u for the U phase, and the asynchronous PWM signal D _uua ^* for the asynchronous PWM control according to this embodiment as an example. The same is true for the V phase and the W phase. FIG. 5(B) shows the carrier signal C _us for the U phase, the correction comparison value Th[n] common to each phase, and the synchronous PWM signal D _uus ^* for the U phase as an example for the synchronous PWM control according to this embodiment. The same is true for the V phase and the W phase.

As shown in FIG. 5A, the carrier signal C _a of the asynchronous PWM control is a triangular wave that repeats peaks and valleys with a period of twice the control period τ (2τ) without depending on the drive frequency (position detection signal θ) of the rotating electric machine 101. In addition, interruption of control calculations such as torque control calculations and updating of parameters according to the results of these control calculations are performed every control period τ. For this reason, for example, the duty command value Duty_u determined as a result of the torque control calculation at time T1 shown by hatching is reflected at time T2. Similarly, the duty command value Duty_u determined as a result of the torque control calculation at time T2 is reflected at time T3. Then, when the carrier signal C _a becomes equal to or greater than the duty command value Duty_u, the asynchronous PWM signal D _uua ^* becomes ON ("1"), and when the carrier signal C _a becomes equal to or less than the duty command value Duty_u, the asynchronous PWM signal D _uua ^* becomes OFF ("0").

As shown in FIG. 5B, the carrier signal _Cus in the synchronous PWM control is a so-called sawtooth wave, which is a composite wave of the position detection signal θ and the voltage phase α ^* in this embodiment. Since the carrier signal _Cus in the synchronous PWM control is determined by the position detection signal θ, the phase of the carrier signal _Cus in the synchronous PWM control does not coincide with the control period τ in principle. However, the interruption of the control calculation such as the torque control calculation and the update of the parameters according to the results of these control calculations are performed every control period τ, so the correction comparison value Th[n] is updated every control period τ. For example, the correction comparison value Th[n] determined as a result of the torque control calculation at time T1 shown by hatching is reflected at time T2. Similarly, the correction comparison value Th[n] determined as a result of the torque control calculation at time T2 is reflected at time T3. In the synchronous PWM control, when the carrier signal _Cus becomes the odd-numbered correction comparison value Th[n] shown by the solid line, the synchronous PWM signal _Duus ^* turns on ("1"), and when the carrier signal _Cus becomes the even-numbered correction comparison value Th[n] shown by the dashed line, the synchronous PWM signal _Duus ^* turns off ("0").

5B, for the sake of simplicity, the correction comparison value Th[n] is depicted as being approximately linear, but the correction comparison value Th[n] changes for each control period τ in response to fluctuations in the drive request for the rotating electric machine 101 and the position-dependent error Δθ _L. In this embodiment, for simplicity, it is assumed that the drive request for the rotating electric machine 101 does not fluctuate and is constant.

6 is an explanatory diagram showing the influence of the position-dependent error Δθ _L included in the position detection signal θ and the action of the correction comparison value Th[n]. FIG. 6(A) shows a synchronous PWM signal _{Duus *} generated by a compare match between a carrier signal _Cus (hereinafter referred to as ideal carrier signal _Cus ^) based on an ideal position detection signal θ not including a position-dependent error Δθ _L and a basic comparison value Th'[n]. FIG. 6(B) shows a synchronous PWM signal Duus ^* generated by a compare match between a carrier signal _Cus (hereinafter referred to as actual carrier signal _Cus or simply carrier signal _Cus ) based on a realistic position detection signal θ including a position-dependent error Δθ L and a basic comparison value Th'[n]. FIG. 6(C) shows a synchronous PWM signal _Duus ^* generated by a compare match between the carrier signal _Cus and the correction comparison value Th[ _n ^] . Although FIG. 6 shows the U-phase as an example, the same applies to the V-phase and W-phase.

6A, when the carrier signal in the synchronous PWM control is the ideal carrier signal C _us ^, the synchronous PWM signal D _us ^* turns on/off at the intersection of the basic comparison value Th'[n] and the ideal carrier signal C _us ^, indicated by the white circle. This is the drive state that should be realized in response to the drive request for the rotating electric machine 101.

However, the real carrier signal _Cus contains a position-dependent error _ΔθL . Furthermore, the real carrier signal _Cus is used directly to generate a carrier signal in the second control device 103 that acquires the position detection signal θ, without going through the first control device 102 that executes arithmetic processing by software. For this reason, the rotating electrical machine control system 100 cannot directly apply correction to the position detection signal θ to correct the position-dependent error _ΔθL .

Therefore, as shown in FIG. 6B, the actual carrier signal C _us has a distorted waveform with respect to the ideal carrier signal C _us ^. As shown by the black circle, the intersection of the actual carrier signal C _us and the basic comparison value Th'[n] shifts in the horizontal time axis direction with respect to the intersection (white circle) in the ideal state. That is, the synchronous PWM signal D _us ^* generated by the compare match of the actual carrier signal C _us and the basic comparison value Th'[n] is shifted from the synchronous PWM signal D _us ^* to be realized in FIG. 6A. Therefore, the driving state according to the driving request for the rotating electric machine 101 is not accurately realized, and the harmonics and ripples of the current flowing through the rotating electric machine 101 may not be sufficiently reduced. Note that the actual carrier signal C _us shown here is shown in a state in which only the first-order error is superimposed on the ideal carrier signal C _us ^ for simplicity.

Therefore, in this embodiment, as shown in FIG. 6C, a synchronous PWM signal D uus * is generated by comparing and matching the actual carrier signal ^C _us and the corrected comparison value Th[n] corrected based on the position-dependent error Δθ _L. The corrected comparison value Th[n] reflects _{the position-dependent error Δθ L that} varies over time according to the rotation of the rotor, so the corrected comparison value Th[n] changes over time according to the position-dependent error Δθ _L. In particular, since the acquisition of the position-dependent error Δθ _L and the calculation of the corrected comparison value Th[n] are performed every control period τ, the corrected comparison value Th[n] changes substantially smoothly according to the variation of the position-dependent error Δθ _L. The corrected comparison value Th[n] shown here shows a state in which a sine wave waveform corresponding to a first-order error is superimposed.

When the correction comparison value Th[n] is used in this way, as shown by the black circle, the intersection between the actual carrier signal C _us and the correction comparison value Th[n] shifts in the phase direction of the vertical axis with respect to the intersection (white circle) in the ideal state. However, the time when the intersection between the actual carrier signal C _us and the correction comparison value Th[n] occurs coincides with the time when the intersection (white circle) occurs in the ideal state. Therefore, the synchronous PWM signal D _uus ^* generated by the compare match between the actual carrier signal C _us and the correction comparison value Th[n] coincides with the synchronous PWM signal D _uus ^* to be realized in FIG. 6(A). As a result, even if the correction by software calculation does not reach the position detection signal θ and the carrier signal C _us is distorted by the position-dependent error Δθ _L , the rotating electric machine 101 is controlled in a substantially ideal driving state in response to the driving request.

As described above, the rotating electric machine control method according to the present embodiment is a rotating electric machine control method that controls the rotating electric machine 101 by the rotating electric machine control system 100 having the first control device 102 that calculates parameters for controlling the driving of the rotating electric machine 101 in response to a drive request for the rotating electric machine 101, and the second control device 103 that executes synchronous PWM control to generate a PWM signal corresponding to the driving frequency of the rotating electric machine 101 using the parameters and a rotor position detection signal θ of the rotating electric machine 101. In this rotating electric machine control method, the second control device 103 acquires the rotor position detection signal θ, and generates carrier signals C _us , C _vs , and C _ws in the synchronous PWM control based on the position detection signal θ. Meanwhile, the first control device 102 acquires a position-dependent error Δθ _L that is an error in rotor position detection and is determined according to the rotor position. Furthermore, in the first control device 102, a basic comparison value Th'[n] for comparison with the carrier signals C _us , C _vs , and C _ws is acquired, and in the first control device 102, a corrected comparison value Th[n] is calculated by correcting the basic comparison value Th'[n] using the position-dependent error Δθ _L. Then, in the second control device 103, synchronous PWM signals D _uus ^* to D _wls ^* are generated by comparing the carrier signals C _us , C _vs , and C _ws with the corrected comparison value Th[n].

When the position detection signal θ is directly used as the carrier signals C _us , C _vs , and C _ws for synchronous PWM control without going through software correction or the like in the first control device 102, the carrier signals C _us , C _vs , and C _ws are distorted in accordance with the detection error of the position detection signal θ. However, in the rotating electric machine control method according to this embodiment, the basic comparison value Th'[n] to be compared with the carrier signals C _us , C _vs , and C _ws to generate the synchronous PWM signals D _uus ^* to D _wls ^* is corrected to a corrected comparison value Th[n] by using the position-dependent error Δθ _L , which is the detection error of the position detection signal θ. As a result, the synchronized PWM signals _Duus ^* to _Dwls * generated by the compare match between the carrier signals _Cus , _Cvs , _Cws and the corrected comparison value Th[n] match the synchronized PWM signals Duus* to _Dwls ^* generated by the compare match between an ideal carrier signal having no detection error in the position detection signal θ and the basic comparison value Th ^' [ ⁿ _] . Therefore, according to the rotating electric machine control method of this embodiment, even when the position detection signal θ contains an error (position-dependent error _ΔθL ), the rotating electric machine 101 is driven in an ideal drive state that should be achieved in response to the drive request, and current harmonics and ripples are reduced.

In particular, the calculation of the corrected comparison value Th[n] and the generation of the synchronous PWM signals D _uus ^* to D _wls ^* using the corrected comparison value Th[n] are performed when at least the modulation factor M of the synchronous PWM signals D _uus ^* to D _wls ^* is greater than a predetermined threshold. That is, the rotating electric machine control method according to this embodiment is performed when the modulation factor M is large, particularly when control is performed in the overmodulation region or square wave region. Synchronous PWM control in the overmodulation region or square wave region where the modulation factor M is large has the advantage of being easier to reduce harmonics and ripples of the current than asynchronous PWM control. However, if the position detection signal θ contains a position-dependent error Δθ _L , even if synchronous PWM control is performed, the harmonics and ripples of the current may not be sufficiently reduced. For this reason, if the rotating electric machine control method according to this embodiment is adopted when performing synchronous PWM control at least in the overmodulation region or square wave region as described above, the harmonics and ripples of the current are more reliably reduced.

Furthermore, in the rotating electric machine control method according to this embodiment, the position-dependent error Δθ _L is acquired and the correction comparison value Th[n] is calculated every predetermined control period τ, thereby changing the correction comparison value Th[n] in accordance with fluctuations in the position-dependent error Δθ _L. In this manner, by updating the position-dependent error Δθ _L and the correction comparison value Th[n] every control period τ, harmonics and ripples in the current caused by the position-dependent error Δθ _L are reduced with particularly good accuracy.

In the rotating electric machine control method according to this embodiment, the position detection signal θ is corrected based on the position-dependent error Δθ _L to generate a corrected position detection signal θ _c , and the position-dependent error Δθ _L is obtained based on this corrected position detection signal θ _c . That is, since the first control device 102 can perform calculation processing according to software, the position-dependent error Δθ _L is not obtained based on the position detection signal θ, which is the original signal, but is obtained based on the corrected position detection signal θ _c . In this way, in the rotating electric machine control method according to this embodiment, the position-dependent error Δθ _L is obtained based on the corrected position detection signal θ _c , so that a particularly accurate position-dependent error Δθ _L can be obtained. Therefore, the accuracy of the calculation of the corrected comparison value Th [n] is improved, and harmonics and ripples of the current are reduced with particularly high precision.

Furthermore, in the rotating electric machine control method according to this embodiment, the corrected position detection signal _θc is further compensated based on the rotor speed N, and the position-dependent error _ΔθL is acquired based on the look-ahead position _θL , which is the compensated corrected position detection signal _θc . In this way, by acquiring the position-dependent error _ΔθL based on the look-ahead position _θL , the deviation caused by the delay time until the result of the torque control calculation, etc. is actually reflected in the rotating electric machine 101 is compensated in advance. Therefore, at the time when the result of the torque control calculation, etc. is actually reflected in the rotating electric machine 101, the harmonics and ripples of the current are reduced with particularly high accuracy.

In addition, in the rotating electric machine control method according to this embodiment, the position-dependent error _ΔθL is determined in advance by the amplitudes A 1 , A 2 , A 2.5 and reference phases δ 1 , δ 2 , δ 2.5 of a first _- order error that varies linearly with respect to the rotor electrical angle (position detection signal θ), a second-order error that varies quadratically with respect to the rotor electrical angle, and/or _a structure-related order error that varies with respect to the rotor electrical angle by an order corresponding to the number of teeth _of the resolver stator. Then, the position-dependent error _ΔθL is obtained by a calculation that combines sine waves or cosine waves based on these amplitudes _{A 1 , A 2 ,} A _2.5 and reference phases _{δ 1} , δ ₂ , δ _2.5 as shown in equation (9).

In this way, when the position-dependent error _ΔθL is obtained by calculation based on the amplitudes A ₁ , A ₂ , A _2.5 and the reference phases δ ₁ , δ ₂ , δ _2.5 , various components that cause the position-dependent error _ΔθL can be comprehensively and efficiently taken into account, and a particularly accurate position-dependent error _ΔθL can be obtained. As a result, current harmonics and ripples are reduced with particularly high precision. In particular, when the structural order error is included in the above calculation, the influence of the components of the position-dependent error ΔθL caused by the mounting accuracy of the resolver 22 (for example, the axial eccentricity of the resolver ₂₂ , the axial offset, the deviation in flatness with respect to the axial end face, and the number of teeth of the resolver stator) can be efficiently corrected.

[Second embodiment]
In the first embodiment, the carrier signals _Cus , _Cvs , and _Cws used in the synchronous PWM control are sawtooth waves, but carrier signals of other forms can be used in the synchronous PWM control. In the second embodiment described below, carrier signals Cus', Cvs _' , and Cws' obtained by folding back the carrier signals _Cus , _Cvs , and _Cws of the synchronous PWM control in the first embodiment at a predetermined phase ( _π ) are used _. Note that parts not mentioned below are configured in the same way as in the first embodiment.

7 is a block diagram showing the configuration of a rotating electric machine control system 200 according to the second embodiment. In the rotating electric machine control system 200, a synchronous PWM signal generator 42 of a second control device 103 generates carrier signals C _us ', C vs ', _and C _ws ' folded back at a phase π in accordance with the following equation (12), based on a position detection signal θ and a voltage phase α ^* .

That is, the carrier signals _Cus ', _Cvs ', and _Cws ' have triangular waveforms obtained by folding back the carrier signals _Cus , _Cvs , and _Cws used in the first embodiment at the intermediate phase value "π". Therefore, the carrier signals _Cus ', _Cvs ', and _Cws ' have a first section in which the carrier signals increase and decrease in proportion to the fluctuation in the rotor position during one period of the rotor electrical angle (position detection signal θ), and a second section in which the carrier signals increase and decrease in inverse proportion to the fluctuation in the rotor position. The first section is a section in which the electrical angle (position detection signal θ) is equal to or greater than zero and less than π, and the second section is a section in which the electrical angle (position detection signal θ) is equal to or greater than π and less than 2π.

7, in the rotating electrical machine control system 200, the synchronous PWM control unit 15 of the first control device 102 receives the read-ahead position _θL and voltage phase α ^* in addition to the voltage norm V _a ^* , the position-dependent error Δθ _L and the battery voltage V _dc . The synchronous PWM control unit 15 then calculates the corrected comparison values Th_u[n], Th_v[n], and Th_w[n] for each of the UVW phases based on these.

Figure 8 is a block diagram showing the configuration of the synchronous PWM control unit 15 in the second embodiment. As shown in Figure 8, in addition to the modulation rate calculation unit 31 and basic comparison value table 32 similar to those in the first embodiment, it also includes a carrier signal estimation unit 51, a correction direction determination unit 52, and a comparison value correction unit 53.

The carrier signal estimation unit 51 estimates the pre-folding carrier signals C _us , C _vs , and C _ws based on the read-ahead position θ _L and the voltage phase α ^* . Specifically, the carrier signal estimation unit 51 includes an adder 61, a first mask processing unit 62 u, a second mask processing unit 62 v, and a third mask processing unit 62 w.

The adder 61 adds the read-ahead position θ _L and the voltage phase α ^* . The first mask processor 62u performs mask processing on the sum (θ _L +α ^* ) of the read-ahead position θ _L and the voltage phase α ^* by modulus calculation with 2π as the divisor. As a result, for the U phase, a carrier signal before turning over at the time when the rotating electric machine 101 is controlled according to the calculation result of the control period τ (hereinafter, referred to as an estimated basic carrier signal C_est_u) is estimated. Similarly, the second mask processor 62v performs mask processing on a value obtained by subtracting 2π/3 from the sum of the read-ahead position θ _L and the voltage phase α ^* by modulus calculation with 2π as the divisor. As a result, for the V phase, an estimated basic carrier signal C_est_v is estimated. Moreover, the third mask processor 62w performs mask processing on a value obtained by adding 2π/3 to the sum of the read-ahead position θ _L and the voltage phase α ^* by modulus calculation with 2π as the divisor. As a result, the estimated fundamental carrier signal C_est_w is estimated for the W phase.

The correction direction determination unit 52 determines the correction direction of the position-dependent error Δθ _{L based on the estimated basic carrier signals C_est_u, C_est_v, and C_est_w of the U, V, and W phases. That is, the correction direction determination unit 52 determines whether the position-dependent error Δθ L should} be added to or subtracted from the basic comparison value Th'[n] based on the estimated basic carrier signals C_est_u, C_est_v, and C_est_w of the U, V, and W phases. The determination result of the correction direction determination unit 52 is represented by the sign of the position-dependent error Δθ _L. In this embodiment, as an example, it is assumed that the position-dependent error Δθ _L is a positive value and the position detection signal θ is greater than the actual position of the rotor.

More specifically, the correction direction determination unit 52 includes a first correction direction determination unit 63u for the U phase, a second correction direction determination unit 63v for the V phase, and a third correction direction determination unit 63w for the W phase. The first correction direction determination unit 63u is composed of, for example, a switch 64 and a multiplier 65. The switch 64 uses the estimated basic carrier signal C_est_u of the U phase as a reference (ref) and switches the signal transmission path between a path that does not pass through the multiplier 65 and a path that passes through the multiplier 65. Specifically, when the electrical angle (position detection signal θ) is smaller than the phase π of the estimated basic carrier signal C_est_u of the U phase, the position-dependent error Δθ _L is output as is. On the other hand, when the electrical angle (position detection signal θ) is equal to or greater than the phase π of the estimated basic carrier signal C_est_u of the U phase, the position-dependent error Δθ _L is multiplied by "-1" by the multiplier 65, and a negative sign is assigned to the position-dependent error Δθ L and output. The same applies to the second correction direction determination unit 63v and the third correction direction determination unit 63w.

That is, in the second embodiment, the correction direction determination unit 52 determines whether the electrical angle (position detection signal θ) is in the first section or the second section of the folded carrier signals C _us ', C _vs ', and C _ws ' based on the estimated basic carrier signals C_est_u, C_est_v, and C_est_w. Then, depending on the result of this determination, the sign of the position-dependent error Δθ _L is inverted depending on the electrical angle. In this embodiment, when the electrical angle is less than π, the value is positive, and when the electrical angle is equal to or greater than π, the value is negative.

The comparison value correction unit 53 calculates corrected comparison values Th_u[n], Th_v[n], and Th_w[n] for each of the U, V, and W phases by correcting the basic comparison value Th'[n], which is set based on the modulation factor M, based on the position-dependent error _ΔθL . Specifically, the comparison value correction unit 53 is composed of a first adder 66u for the U phase, a second adder 66v for the V phase, and a third adder 66w for the W phase. The first adder 66u calculates the corrected comparison value Th_u[n] for the U phase by adding the position-dependent error _ΔθL input via the first correction direction determination unit 63u to the basic comparison value Th'[n]. Due to the action of the first correction direction determination unit 63u, the position-dependent error _ΔθL switches between positive and negative depending on the phase, so the value obtained by adding or subtracting the position-dependent error _ΔθL to or from the basic comparison value Th'[n] depending on the phase becomes the corrected comparison value Th_u[n] for the U phase.

Therefore, when the folded carrier signal _Cus ' of the U phase is in the first section (section from 0 to π), the basic comparison value Th'[n] is changed in a predetermined direction according to the position-dependent error _ΔθL , thereby calculating the corrected comparison value Th_u[n] of the U phase. On the other hand, when the folded carrier signal Cus' of the U phase is in the second section (section from π to 2π before folding), the basic comparison value Th'[n] is changed in a direction opposite to the predetermined direction according to the position-dependent error _ΔθL , thereby calculating the corrected comparison value Th_u[n] of the U phase. The "predetermined direction" is a direction in which the basic comparison value Th'[n] is increased or decreased, and is determined according to whether the carrier signal Cus' is in the first section or the second section, and the positive or negative of the position-dependent error _ΔθL . In this embodiment, when the carrier signal Cus' is in the first section, the basic comparison value Th'[n] is increased, and when the carrier signal Cus' is in the second section, the basic comparison value Th'[n] is decreased.

Similarly, the second adder 66v calculates a corrected comparison value Th_v[n] for the V phase by adding the position-dependent error _ΔθL input via the second correction direction determination unit 63v to the basic comparison value Th'[n]. The third adder 662 calculates a corrected comparison value Th_w[n] for the V phase by adding the position-dependent error _ΔθL input via the third correction direction determination unit 63w to the basic comparison value Th'[n].

Then, the synchronous PWM signal generator 42 of the second control device 103 generates synchronous PWM signals _{D uus} *, D _uls ^* for the U-phase switching elements by comparing and matching the U-phase carrier signal C us ' folded back at phase ^π and the U-phase correction comparison value Th_u[ _n ].

Similarly, the synchronous PWM signal generator 42 generates synchronous PWM signals D vus *, D vls * for the V-phase switching element by a compare match between the V-phase carrier signal C _vs ' folded back at phase π and the V-phase correction comparison value Th_v[n]. The synchronous PWM signal generator 42 also generates synchronous PWM signals D _wus ^* , D _wls ^* for the W-phase switching element by a compare match between the W-phase carrier signal C _ws ' folded back at phase _π ^and the V-phase correction comparison value Th_w[ _n ^] .

9 is a timing chart showing the change over time of the carrier signal and the synchronous PWM signal in the second embodiment. As shown in FIG. 9, in the second embodiment, the carrier signal C _us ' has a waveform obtained by folding back the U-phase carrier signal C _us at phase π. Therefore, the basic comparison value Th' [n] and the corrected comparison value Th_u [n] are about half the number (n = 1 to 6) of the first embodiment, and the synchronous PWM signal D _uus ^* similar to that in the first embodiment can be generated. Here, the U phase is taken as an example, but the same applies to the V phase and the W phase.

10 is an explanatory diagram showing the operation of the rotating electric machine control system 200 of the second embodiment. Fig. 10(A) is a graph showing the relationship between the ideal carrier signal C _us ^ of the U phase without folding back and the actual carrier signal C _us including an error, and the correction comparison values Th[1] and Th[2] of the first embodiment, for comparison. Fig. 10(B) is a graph showing the relationship between the ideal carrier signal C _us ^' of the folded back U phase and the actual carrier signal C _us ', and the correction comparison value Th[1] of the first embodiment in which the correction direction does not change depending on the phase. Fig. 10(C) is a graph showing the relationship between the ideal carrier signal C _us ^' of the folded back U phase and the folded back actual carrier signal C _us ', and the correction comparison value Th_u[1] in which the correction direction is changed depending on the phase.

In Fig. 10, time t1 is the time when the synchronous PWM signal D _uus ^* should be turned on when the rotating electric machine 101 is driven in response to a drive request. Time t2 is the time when the phases of the carrier signals C _us ^, C _us , etc. are π. Time t3 is the time when the synchronous PWM signal D _uus ^* should be turned off when the rotating electric machine 101 is driven in response to a drive request. Black circles indicate the intersections of each corrected comparison value and each carrier signal, and white circles indicate the intersections of the ideal carrier signal C _us ^ or C _us ^' and the basic comparison values Th'[1], Th'[2]. Here, the U phase is taken as an example, but the same applies to the V phase and the W phase.

10A, when a non-wrapped carrier signal _Cus is used in synchronous PWM control, the carrier signal _Cus and the corrected comparison values Th[1] and Th[2] intersect at ideal times t1 and t3 according to the drive request. Therefore, when a non-wrapped carrier signal _Cus is used, the synchronous PWM signal _Duus ^* is turned on/off at ideal timing according to the drive request by the compare match of the corrected comparison values Th[1] and Th[2].

On the other hand, as shown in FIG. 10B, when the folded carrier signal C _us ' is used in the synchronous PWM control, the intersection point of the carrier signal C _us ' and the correction comparison value Th [1] may be shifted in the time axis direction. Specifically, in the first section (0 to π section) before folding, the timing (black circle) at which the carrier signal C _us ' and the correction comparison value Th [1] cross each other coincides with the timing (white circle) at which the ideal carrier signal C _us ^' and the basic comparison value Th' [1] cross each other. However, in the second section after folding (π to 2π section before folding), even though the correction comparison value Th [1] reflecting the position-dependent error Δθ _L is used, the timing (black circle) at which the carrier signal C _us ' and the correction comparison value Th [1] cross each other is shifted in the time axis direction from the timing (white circle) at which the ideal carrier signal C _us ^' and the basic comparison value Th' [1] cross each other. Therefore, the synchronous PWM signal _Duus ^* is not turned off at an ideal timing according to a drive request.

Therefore, in the second embodiment, when the folded carrier signal C _us ' is used in the synchronous PWM control, it is determined whether the carrier signal C _us ' is in the first section or the second section. Then, the positive or negative of the position-dependent error Δθ _L is changed depending on whether the carrier signal C _us ' is in the first section or the second section, thereby adjusting the correction direction. This corresponds to inverting the correction comparison value Th_u[1] in the direction of the vertical axis in the second section from time t2 to time t3, as shown in FIG. 10(C). As a result, even in the second section, the timing (black circle) at which the carrier signal C _us ' and the correction comparison value Th[1] cross each other coincides with the timing (white circle) at which the ideal carrier signal C _us ' and the basic comparison value Th'[1] cross each other. As a result, even when the folded carrier signal C _us ' is used in synchronous PWM control, the synchronous PWM signal D _us ^* turns on/off at ideal timing according to the drive request.

As described above, in the second embodiment, carrier signals C us ', C vs ', C ws ' are generated having a first section that increases or decreases in proportion to fluctuations in the rotor position and a second section that increases or decreases in inverse proportion to fluctuations in the rotor position during one period of the rotor electrical angle ₍ position detection signal θ), and having waveforms that are folded back between the first section _and the second _section . Therefore, in the rotating electric machine control method according to the second embodiment, it is determined whether the electrical angle (position detection signal θ) is in the first section or the second section of the carrier signals C _us ', C _vs ', C _ws '. When the electrical angle (position detection signal θ) is in a first interval of the carrier signals _Cus ', _Cvs ', and _Cws ', the corrected comparison values Th_u[ _n ], Th_v[n], and Th_w[n] are calculated by changing the basic comparison value Th'[n] in a predetermined direction in accordance with the position-dependent error Δθ _L. On the other hand, when the electrical angle (position detection signal θ) is in a second interval of the carrier signals _Cus ', _Cvs ', and _Cws ', the corrected comparison values Th_u[n], Th_v[n], and Th_w[n] are calculated by changing the basic comparison value Th'[n] in a direction opposite to the predetermined direction in accordance with the position-dependent error Δθ L.

In this way, when folded carrier signals C _us ', C _vs ', C _ws ' in synchronous PWM control are used, the first and second sections are determined, and the correction direction is adjusted according to the result. As a result, even when folded carrier signals C _us ', C _vs ', C _ws ' are used, synchronous PWM signal D _uus ^* is turned on/off at ideal timing according to drive requirements, thereby reducing harmonics and ripples in the current.

In the second embodiment, the control is exemplified in the case where the position-dependent error _ΔθL is a positive value and the position detection signal θ is greater than the actual position of the rotor, so the basic comparison value Th'[n] is increased in the first interval and decreased in the second interval. Therefore, in practice, when the position-dependent error _ΔθL is a negative value, the basic comparison value Th'[n] is decreased in the first interval and increased in the second interval.

In addition, in the rotating electric power control method of the second embodiment, the correction comparison value Th_u[n] is calculated for each UVW phase, so the rotating electric power control method of the second embodiment has the advantage of improving the degree of freedom of dead time compensation related to the switching of the inverter 104 during synchronous PWM control. In this way, the degree of freedom of dead time compensation is improved, making it possible to perform highly accurate dead time compensation, which in particular reduces harmonics and ripples in the current.

In addition, when the folded carrier signals C _us ', C _vs ', and C _{ws '} are used as in the second embodiment, the number of the basic comparison value Th _' [ _n ] and the corrected comparison values Th_u [n], Th_v [n], and Th_w [n] of each phase is smaller than the case of using the carrier signals C us , C vs , and C ws that are not folded. Therefore, the storage area for holding the basic comparison value Th' [n] is saved, and the calculation load for obtaining the corrected comparison values Th_u [n], Th_v [n], and Th_w [n] is also suppressed. In the second embodiment, the carrier signal is folded once at the phase π, but the carrier signal may be folded, for example, every 2π/3 or every 2π/4. The effect of saving the storage area and reducing the calculation load becomes more noticeable as the number of times the carrier signal is folded is increased.

In the second embodiment, the carrier signal estimation unit 51 estimates the carrier signals _Cus , _Cvs , and _Cws before folding in the synchronous PWM control unit 15 of the first control unit 102. However, when the first control unit 102 (synchronous PWM control unit 15) can monitor the carrier signals _Cus ', _Cvs ', and _Cws ' of the synchronous PWM control by, for example, acquiring them from the second control unit 103 (synchronous PWM signal generator 42), the carrier signal estimation unit 51 may be omitted and the correction direction may be adjusted according to the monitored carrier signals _Cus ', _Cvs ', and _Cws '.

[Third embodiment]
In the first embodiment, the sum (θ+α ^* ) of the position detection signal θ and the voltage phase α ^* is used as the carrier signals _Cus , _Cvs , and _Cws of the synchronous PWM control, and in the second embodiment, the folded carrier signals _Cus ', _Cvs ', and _Cws ' are used for the synchronous PWM control. That is, in the first and second embodiments, apart from the presence or absence of folding, the signal obtained by offsetting the position detection signal θ with the voltage phase α ^* is used as the carrier signal for the synchronous PWM control. However, in the synchronous PWM control, the position detection signal θ can be used as the carrier signal as it is. In the third embodiment, a form in which the position detection signal θ is used as the carrier signal as it is will be described below. Note that the parts not mentioned below are configured in the same manner as in the first embodiment.

11 is a block diagram showing a configuration of a rotating electrical machine control system 300 according to the third embodiment. In the rotating electrical machine control system 300, a synchronous PWM signal generator 42 of a second control device 103 generates carrier signals C _us , C _vs , and C _ws based on a position detection signal θ in accordance with the following equation (13).

As shown in FIG. 11, in the rotating electrical machine control system 300, the voltage phase α ^* is not input to the synchronous PWM signal generator 42 of the second control device 103, ^but is input to the synchronous PWM control unit 15.

Fig. 12 is a block diagram showing the configuration of the synchronous PWM control unit 15 in the third embodiment. As shown in Fig. 12, the synchronous PWM control unit 15 further includes an adder 71 in addition to the adder 34 and mask processing unit 35 in the comparison value correction unit 33. The adder 71 adds the voltage phase α ^* to the position-dependent error _ΔθL . Therefore, in the third embodiment, a change in the voltage phase α ^* is reflected in the corrected comparison value Th[n].

As in the above-described rotating electric machine control system 300, when the position detection signal θ is used as is for the carrier signals C _us , C _vs , and C _ws , and the voltage phase α ^* is reflected in the corrected comparison value Th[n], the same effects as those of the rotating electric machine control system 100 of the first embodiment can be achieved.

In the third embodiment, the position detection signal θ is used as is for the carrier signals C _us , C _vs , and C _ws. However, while the position detection signal θ is used for the carrier signals C _us , C _vs , and C _ws , a carrier signal folded back as necessary may be used for synchronous PWM control as in the second embodiment.

[Fourth embodiment]
In the above first to third embodiments, the error Δθ relating to the detection of the position detection signal θ is obtained by calculation according to the formula (5) and the formula (6), but this is not limiting. As in the fourth embodiment described below, the error Δθ relating to the detection of the position detection signal θ may be obtained by referring to a lookup table.

Figure 13 is a block diagram showing the configuration for the error correction unit 17 of the fourth embodiment to calculate the error Δθ related to position detection. As shown in Figure 13, in this fourth embodiment, the first control device 102 stores the rotor position index array x[n] and the error array y[n] in advance in a memory or the like. Then, the error correction unit 17 obtains the error Δθ related to the detection of the position detection signal θ using the position detection signal θ, the rotor position index array x[n], and the error array y[n].

The rotor position index array x[n] is one-dimensional array data that defines a position index according to the electrical angle of the rotor. The error array y[n] is one-dimensional array data that stores position detection errors y ₁ to y _n that occur when the rotor is in a position represented by a specific electrical angle, in correspondence with the rotor position index array x[n]. More specifically, the error array y[n] represents a first-order error, a second-order error, and/or a structure-related order error. The data of the error array y[n] (errors y ₁ to y _n ) is determined in advance by experiments, numerical calculations, etc.

Figure 14 is a graph showing the relationship between the rotor position index array x[n] and the error array y[n]. For simplicity, Figure 14 shows the relationship between the rotor position index array x[n] and the error array y[n] when only a first-order error occurs in the rotor position detection. When a first-order error occurs in the rotor position detection, as shown in Figure 14, the relationship between the rotor position index array x[n] and the error array y[n] is, for example, a sine wave.

Figure 15 is a graph showing the relationship between the true rotor position and the error Δθ obtained by directly referencing the rotor position index array x[n] and the error array y[n] based on the position detection signal θ. As shown by the solid line in Figure 15, if only a first-order error occurs in rotor position detection, the error Δθ should be, for example, a sine wave for the true rotor position. However, as shown by the dashed line in Figure 15, if the error Δθ is obtained by directly referencing the rotor position index array x[n] and the error array y[n] based on the position detection signal θ, which is the actual detection value, the obtained error Δθ will be a distorted waveform from the original sine wave waveform for the true rotor position. This is because the actual position detection signal θ contains the error Δθ, and a shift occurs in the reference position of the rotor position index array x[n] by the amount of the error Δθ.

Therefore, the error correction unit 17 in the fourth embodiment is configured to obtain the correct error Δθ for the true rotor position using the position detection signal θ, the rotor position index array x[n], and the error array y[n]. Specifically, the error correction unit 17 includes a subtraction unit 81 and a LUT 82, which is a one-dimensional lookup table (see FIG. 13).

The subtraction unit 81 calculates a new one-dimensional array, the difference array (x[n] - y[n]), by subtracting the error array y[n] from the rotor position index array x[n]. The LUT 82 is configured to use the difference array (x[n] - y[n]) as the input reference array and the error array y[n] as the output reference array.

Figure 16 is a graph showing the relationship between the difference array (x[n] - y[n]) and the error array y[n]. As shown in Figure 16, in the LUT 82, the difference array (x[n] - y[n]) is the input reference array, while the error array y[n] is the output reference array, it has a waveform that is distorted from a sine wave. This waveform distortion corresponds to the deviation between the solid and dashed lines in Figure 15. Therefore, when the error Δθ associated with the detection of the position detection signal θ is obtained by referring to the LUT 82 based on the position detection signal θ, this error Δθ becomes a correct value with respect to the true rotor position.

In the fourth embodiment, when the error correction unit 17 calculates the corrected position detection signal _θc , the error Δθ relating to the detection of the position detection signal θ is obtained by the LUT 82. Similarly, the error generation unit 20 can be configured to obtain the position-dependent error _ΔθL by referring to the LUT 82.

Specifically, in the first to third embodiments, the look-ahead compensation unit 19 calculates the look-ahead position θ _L using the corrected position detection signal θ _c , and the error generation unit 20 calculates the position-dependent error Δθ _L according to equation (9) based on the look-ahead position θ _L. However, the look-ahead compensation unit 19 may be configured to receive the position detection signal θ before correction. In this case, the error generation unit 20 obtains the position-dependent error Δθ L based on the look-ahead position θ _L calculated based on the position detection signal θ before correction, that is, the look-ahead position θ _L including the error Δθ related to the position detection. For this reason, if the error generation unit 20 obtains the position-dependent error Δθ _L by directly referring to the rotor position index array x[n] and the error array y[n], there are cases where the error generation unit 20 cannot obtain an accurate position-dependent error _{Δθ L for} the true rotor position.

Therefore, when calculating the position-dependent error Δθ L based on the read-ahead position θ _L calculated based on the position detection signal _θ before correction, it is preferable that the error generating unit 20 calculates the position-dependent error Δθ _L similarly to the above-mentioned fourth embodiment, instead of calculating the formula (9). That is, it is preferable that the error generating unit 20 calculates the position-dependent error Δθ _L from the read-ahead position θ _L by the subtraction unit 81 and the LUT 82 using the rotor position index array x[n] and the error array y[n] similarly to the above-mentioned fourth embodiment. By acquiring the position-dependent error Δθ _L in this way, even when calculating the position-dependent error Δθ _L based on the read-ahead position θ _L calculated based on the position detection signal θ before correction, the error generating unit 20 can acquire an accurate position-dependent error Δθ _L for the true rotor position.

In the fourth embodiment, an example is shown in which the error Δθ and the position-dependent error Δθ _L include only a first-order error, but the case in which a second-order error and a structure-attributable order error are included is also similar to the fourth embodiment.

In the fourth embodiment, the LUT 82 is configured to use the difference array (x[n]-y[n]) obtained by correcting the rotor position index array x[n] with the error array y[n] as the input reference array, and the error array y[n] as the output reference array, but this is not limited to the above. The LUT 82 may be configured to use the rotor position index array x[n] as the input reference array, and use the array (e.g., y[n]-x[n]) obtained by correcting the rotor position index array x[n] with the error array y[n] as the output reference array. In this case, too, an accurate error Δθ and a position-dependent error Δθ _L are obtained for the true rotor position, as in the fourth embodiment.

As described above, in the rotating electric machine control method according to the fourth embodiment, the error Δθ or the position-dependent error Δθ _L is determined in advance by the first array data (rotor position index array x[n]) that determines the position index according to the electrical angle of the rotor, and the second array data (error array y[n]) that determines the first-order error that varies linearly with respect to the electrical angle of the rotor (position detection signal θ), the second-order error that varies quadratically with respect to the electrical angle of the rotor, and/or the structure-related order error that varies with respect to the electrical angle of the rotor, with an order corresponding to the number of teeth in the stator of the resolver 22 that detects the position of the rotor. The error Δθ or the position-dependent error Δθ _L is acquired by referring to the LUT 82 that uses the first array data (rotor position index array x[n]) corrected by the second array data (error array y[n]) as the input reference array and uses the second array data (error array y[n]) as the output reference array. Alternatively, the error Δθ or the position-dependent error Δθ _L is obtained by referring to an LUT 82 that uses the first array data (rotor position index array x[n]) as an input reference array and uses the second array data (error array y[n]) corrected by the first array data (rotor position index array x[n]) as an output reference array.

In this way, when the error Δθ or the position-dependent error _ΔθL is obtained using the LUT 82, a particularly accurate error Δθ and position-dependent error _ΔθL are obtained with respect to the true rotor position. As a result, even when the position detection signal θ contains an error (position-dependent error _ΔθL ), the rotating electric machine 101 is driven in an ideal drive state that should be achieved in response to a drive request, and current harmonics and ripples are particularly reduced.

The above describes the embodiments of the present invention, but the configurations described in the above embodiments and each modified example are merely examples of application of the present invention and are not intended to limit the technical scope of the present invention.

11: Torque control unit, 13: Asynchronous PWM control unit, 14: Vector conversion unit, 15: Synchronous PWM control unit, 17: Error correction unit, 19: Look-ahead compensation unit, 20: Error generation unit, 22: Resolver, 32: Basic comparison value table, 33: Comparison value correction unit, 41: Asynchronous PWM signal generator, 42: Synchronous PWM signal generator, 43: PWM output switch, 51: Carrier signal estimation unit, 52: Correction direction determination unit, 53: Comparison value correction unit, 100: Rotating electric machine control system, 101: Rotating electric machine, 102: First control unit, 103: Second control unit, 104: Inverter, 105: Battery, 200: Rotating electric machine control system, 300: Rotating electric machine control system

Claims (8)

A rotating electric machine control method for controlling a rotating electric machine by a rotating electric machine control system having a first control device that calculates parameters for controlling a drive of the rotating electric machine in response to a drive request for the rotating electric machine, and a second control device that executes synchronous PWM control to generate a PWM signal according to a drive frequency of the rotating electric machine by using the parameters and a position detection signal of a rotor of the rotating electric machine, the method comprising:
The second control device acquires the position detection signal of the rotor;
a carrier signal for the synchronous PWM control is generated in the second control device based on the position detection signal;
In the first control device, a position-dependent error is obtained, which is an error in detecting the position of the rotor and is determined depending on the position of the rotor;
obtaining, in the first control device, a base comparison value for comparing with the carrier signal;
calculating a corrected comparison value by correcting the basic comparison value using the position-dependent error in the first control device;
In the second control device, the carrier signal is compared with the correction comparison value to generate the PWM signal;
The calculation of the correction comparison value and the generation of the PWM signal using the correction comparison value are executed when a modulation rate of the PWM signal is greater than a predetermined threshold value.
A rotating electrical machine control method. A method for controlling a rotating electric machine according to claim 1 , comprising the steps of:
The acquisition of the position-dependent error and the calculation of the correction comparison value are performed at every predetermined control period, thereby changing the correction comparison value in accordance with a fluctuation of the position-dependent error.
A rotating electrical machine control method. 3. A rotating electric machine control method according to claim 1, further comprising:
generating the carrier signal having a waveform that is folded back between a first section that increases or decreases in proportion to a fluctuation in the position of the rotor and a second section that increases or decreases in inverse proportion to the fluctuation in the position of the rotor during one period of the electrical angle of the rotor;
determining whether the electrical angle is in the first section or the second section of the folded back carrier signal;
when the electrical angle is in the first section of the folded back carrier signal, the basic comparison value is changed in a predetermined direction in response to the position-dependent error to calculate the corrected comparison value;
when the electrical angle is in the second section of the folded back carrier signal, the basic comparison value is changed in a direction opposite to the predetermined direction in response to the position dependent error, thereby calculating the corrected comparison value.
A rotating electrical machine control method. A rotating electric machine control method according to any one of claims 1 to 3 ,
generating a corrected position detection signal by correcting the position detection signal based on the position dependent error;
The position dependent error is obtained based on the compensated position detection signal.
A rotating electrical machine control method. A rotating electrical machine control method according to claim 4 , comprising:
Further compensating the corrected position detection signal based on the rotation speed of the rotor;
The position-dependent error is obtained based on the compensated position detection signal.
A rotating electrical machine control method. A rotating electric machine control method according to any one of claims 1 to 5 ,
the position-dependent error is determined in advance by an amplitude and a reference phase of an error that varies linearly with respect to the electrical angle of the rotor, an error that varies quadratically with respect to the electrical angle of the rotor, and/or an error that varies with respect to the electrical angle of the rotor with an order corresponding to the number of teeth in a stator of a resolver that detects the position of the rotor;
The position-dependent error is obtained by a calculation of synthesizing a sine wave or a cosine wave based on the amplitude and the reference phase.
A rotating electrical machine control method. A rotating electric machine control method according to any one of claims 1 to 5 ,
the position-dependent error is predetermined by first array data that defines a position index based on the electrical angle of the rotor, and second array data that defines an error that varies linearly with respect to the electrical angle of the rotor, an error that varies quadratically with respect to the electrical angle of the rotor, and/or an error that varies with respect to the electrical angle of the rotor with an order corresponding to a number of teeth in a stator of a resolver that detects the position of the rotor,
The position dependent error is
The first array data corrected by the second array data is acquired by referring to a lookup table in which the first array data is an input reference array and the second array data is an output reference array; or
The first array data is acquired by referring to a lookup table in which the first array data is an input reference array and the second array data corrected by the first array data is an output reference array.
A rotating electrical machine control method. A rotating electric machine control system including: a first control device that calculates parameters for controlling a drive of a rotating electric machine in response to a drive request for the rotating electric machine; and a second control device that executes synchronous PWM control to generate a PWM signal according to a drive frequency of the rotating electric machine by using the parameters and a position detection signal of a rotor of the rotating electric machine,
The second control device is
a position detection signal acquirer for acquiring the position detection signal of the rotor;
a carrier signal generator that generates a carrier signal in the synchronous PWM control based on the position detection signal;
Equipped with
The first control device is
a position-dependent error acquisition unit that acquires a position-dependent error, which is an error in detecting the position of the rotor and is determined depending on the position of the rotor;
a comparison value acquisition unit for acquiring a basic comparison value to be compared with the carrier signal;
a comparison value correction unit that calculates a corrected comparison value by correcting the basic comparison value using the position dependent error;
Equipped with
The second control device further includes a PWM signal generator that generates the PWM signal by comparing the carrier signal with the correction comparison value ,
The calculation of the correction comparison value and the generation of the PWM signal using the correction comparison value are executed when a modulation rate of the PWM signal is greater than a predetermined threshold value.
Rotating electric machine control system.