Electric motor control method and electric motor control device

Abstract

This electric motor control method compensates for torque ripple by correcting a torque command value. In this electric motor control method, for each of a plurality of orders, the fundamental amplitude of torque ripple is calculated, and a correction amplitude of each order is calculated by correcting the fundamental amplitude of each order in accordance with the rotational speed of the electric motor. Then, a compensation torque is calculated using the correction amplitude of each order, and the torque command value is corrected using this compensation torque.

Description

Electric motor control method and electric motor control device

[0001] The present invention relates to a control method and control device for an electric motor.

[0002] JP2001-197765A discloses a servo motor drive system that feeds back torque corresponding to torque pulsation to the torque command according to the motor rotation angle. This system calculates the amount of torque ripple and the vibration frequency from the torque command using FFT (Fast Fourier Transform).

[0003] In electric motor control, torque ripple is sometimes compensated by correcting the torque command value. In principle, torque ripple can be compensated by correcting the torque command value only when the corrected torque command value is equal to the maximum torque the motor can output (hereinafter referred to as maximum output torque T). ｍａｘ This refers to the following cases: And, the torque command value is large, and the maximum output torque T ｍａｘ If the value is close to the maximum output torque T, the corrected torque command value will be the maximum output torque T. ｍａｘ This can lead to limitations. In this case, torque ripple may not be adequately compensated.

[0004] Furthermore, torque ripple is a harmonic that corresponds to the rotational speed of the motor, and the torque output by the motor has multiple torque ripples of different orders superimposed on it, depending on the specific configuration (number of phases, etc.) and rotational speed of the motor. Therefore, torque ripple compensation is performed for multiple orders. As mentioned above, the corrected torque command value is the maximum output torque T. ｍａｘ The situation in which this is restricted is equivalent to a situation in which compensation for torque ripple of each order is uniformly restricted.

[0005] This invention relates to a torque command value that corresponds to the maximum output torque T ｍａｘ The objective is to provide an electric motor control method and an electric motor control device that can compensate for torque ripple better than conventional methods when the torque ripple is a large value close to [a certain value].

[0006] One aspect of the present invention is a motor control method that compensates for torque ripple by correcting the torque command value. In this motor control method, the fundamental amplitude of the torque ripple is calculated for each of several orders, and the corrected amplitude for each order is calculated by correcting the fundamental amplitude for each order according to the rotational speed of the motor. Then, a compensating torque is calculated using the corrected amplitude for each order, and the torque command value is corrected using this compensating torque.

[0007] Figure 1 is a block diagram showing the schematic configuration of an electric vehicle. Figure 2 is a block diagram showing the configuration of the controller. Figure 3 is a block diagram showing the configuration of the torque ripple calculation unit. Figure 4 is a graph schematically showing the amplitude of torque ripple with respect to the torque command value. Figure 5 is an explanatory diagram showing the setting of priority. Figure 6 is a graph schematically showing the phase of torque ripple with respect to the torque command value. Figure 7 is a flowchart related to torque ripple compensation. Figure 8 is a flowchart related to the calculation of the correction amplitude. Figure 9 is an explanatory diagram showing the compensation torque in a comparative example. Figure 10 is an explanatory diagram showing the compensation torque of this embodiment. Figure 11 is a graph showing the operation of a comparative example. Figure 12 is a graph showing the operation of this embodiment.

[0008] Embodiments of the present invention will be described below with reference to the drawings.

[0009] Figure 1 is a block diagram showing the schematic configuration of the electric vehicle 100. As shown in Figure 1, it includes a battery 10, a powertrain 11, and a controller 12.

[0010] Battery 10 is a DC power source that stores power supplied to the powertrain 11. Battery 10 is composed of, for example, a lithium-ion battery and is rechargeable. The DC voltage V output by battery 10 ｄｃ This can be acquired or measured as appropriate by a current sensor or the like (not shown).

[0011] The powertrain 11 is a component that transmits power to the drive wheels (not shown). In this embodiment, the powertrain 11 consists of an inverter 13 (INV), an electric motor 14, and a drive shaft 15, etc.

[0012] The inverter 13 converts the DC power input from the battery 10 into AC power and supplies it to the electric motor 14. When the electric motor 14 is rotated while being driven by the drive wheels, the inverter 13 converts the AC power (so-called regenerative power) generated by the electric motor 14 into DC power and inputs it to the battery 10.

[0013] The electric motor 14 is a drive source of the electric vehicle 100. The output shaft of the electric motor 14 is connected to the drive wheels via a speed reducer and a drive shaft 15 not shown. Therefore, the torque of the electric motor 14 generates a driving force on the drive wheels. The rotation angle θ [rad] and the rotation speed ω [rad / s] of the electric motor 14 can be appropriately acquired using a rotation detector such as an encoder not shown. In the present embodiment, the rotation angle θ is the electrical angle ψ ｅ and the rotation speed ω is the electrical angular velocity ω ｅ . As a parameter representing the rotation angle, instead of the electrical angle ψ ｅ , the mechanical angle ψ ｍ can be used. Further, as a parameter representing the rotation speed, instead of the electrical angular velocity ω ｅ , the mechanical angular velocity ω ｍ [rad / s] or the rotation speed [rpm] can be used.

[0014] In the present embodiment, the electric motor 14 is, for example, a three-phase AC synchronous motor. Therefore, the ripple (hereinafter referred to as torque ripple) superimposed on the torque of the electric motor 14 appears as a harmonic having an order that is a multiple of 6 according to the rotation speed of the electric motor 14. That is, the order n of the torque ripple is represented by n = 6m (m = 1, 2, 3,...). In the present embodiment, a plurality of orders n of torque ripple are compensation targets. However, depending on the rotation speed ω of the electric motor 14, some of the orders n of torque ripple among the plurality of orders n of torque ripple to be compensated may be selectively (preferably) compensated. For example, only a specific one of the orders n of torque ripple among the plurality of orders n of torque ripple may be selectively (preferably) compensated.

[0015] Incidentally, the current flowing through each of the U, V, and W phases of the electric motor 14 (hereinafter, phase current i [[ID=2�]] ＵＶＷThis can be acquired as needed by a current sensor (not shown).

[0016] The controller 12 is a control device that comprehensively controls the operation of various parts of the electric vehicle 100, such as the battery 10 and the powertrain 11. The controller 12 is composed of, for example, one or more computers and is programmed to control the operation of each part at a predetermined control cycle.

[0017] For example, the controller 12 controls the accelerator pedal input amount A. ｐｏ , the DC voltage V of battery 10 ｄｃ Furthermore, the controller controls the torque of the electric motor 14 based on the rotation angle θ and rotation speed ω of the electric motor 14. In other words, the controller 12 functions as a control device for the electric motor 14. Specifically, the controller 12 controls the accelerator operation amount A ｐｏ , the DC voltage V of battery 10 ｄｃ Furthermore, based on the rotation angle θ and rotation speed ω of the electric motor 14, the controller 12 generates a PWM (Pulse Width Modulation) signal to control the switching operation of the inverter 13. By operating the inverter 13 according to this PWM signal, the controller 12 controls the accelerator operation amount A by the electric motor 14. ｐｏ It outputs torque according to the conditions.

[0018] Figure 2 is a block diagram showing the configuration of the controller 12. Here, the configuration directly related to the torque ripple compensation process (hereinafter referred to as torque ripple compensation) is shown, and other configurations related to vibration damping control, etc., are not shown or described. As shown in Figure 2, the controller 12 includes a torque command value calculation unit 20, a torque ripple calculation unit 21, a torque command value correction unit 22, and a current control unit 23.

[0019] The torque command value calculation unit 20 calculates the accelerator operation amount A ｐｏ Based on the rotational speed ω of the electric motor 14, a torque command value (hereinafter referred to as the first torque command value T) is determined that represents the torque that the electric motor 14 should output. １ ＊ The first torque command value T is calculated. １ ＊This is a basic torque command value determined according to the operation of the accelerator, and represents the torque requested by the driver to the electric vehicle 100 (electric motor 14). In this embodiment, the torque command value calculation unit 20 calculates the accelerator operation amount A ｐｏ and rotational speed ω and first torque command value T １ ＊ It has a torque map (not shown) that associates with and . Therefore, the torque command value calculation unit 20 refers to this torque map and calculates the accelerator operation amount A ｐｏ and the first torque command value T corresponding to the rotational speed ω. １ ＊ Perform the calculation.

[0020] The torque ripple calculation unit 21 calculates the first torque command value T １ ＊ The torque ripple is calculated based on the rotational speed ω and the rotational angle θ. In this embodiment, the torque ripple calculation unit 21 calculates the torque ripple of each of the multiple orders n to be compensated. The torque ripple calculation unit 21 then sums up the torque ripples of each order n to obtain the total amount of torque ripple to be compensated (hereinafter referred to as the compensated torque T). Σ Perform the calculation ′.

[0021] The torque command value correction unit 22 compensates the torque T Σ Using ′, the first torque command value T １ ＊ This corrects the torque command value. As a result, the torque command value correction unit 22 corrects the torque command value (hereinafter referred to as the second torque command value T) so that the torque ripple is compensated. ２ ＊ In this embodiment, the torque command value correction unit 22 is a subtractor, and calculates the first torque command value T. １ ＊ Compensation Torque T Σ By subtracting ', the second torque command value T ２ ＊ Perform the calculation.

[0022] The current control unit 23 sets the second torque command value T ２ ＊ Based on this, a PWM signal is generated. More specifically, the current control unit 23 generates a second torque command value T ２＊ DC voltage V ｄｃ , rotational speed ω, rotational angle θ, and phase current i ＵＶＷ Based on the above, the phase current i ＵＶＷ The current control unit 23 calculates the target value (command value) for the phase current i. ＵＶＷ The PWM signal is generated so that it matches or follows this target value.

[0023] Figure 3 is a block diagram showing the configuration of the torque ripple calculation unit 21. As shown in Figure 3, the torque ripple calculation unit 21 includes a basic amplitude calculation unit 31, a correctable amount calculation unit 32, a filter processing unit 33, a priority setting unit 34, an amplitude correction unit 35, a basic wave calculation unit 36, and a compensated torque calculation unit 37.

[0024] The basic amplitude calculation unit 31 calculates the first torque command value T １ ＊ Based on this, the amplitude of the torque ripple is calculated. The basic amplitude calculation unit 31 calculates the amplitude for each of the multiple order n torque ripples to be compensated. Hereinafter, the amplitude of each order n calculated by the basic amplitude calculation unit 31 is referred to as the basic amplitude K ｎ That's what they say.

[0025] Basic amplitude K ｎ In other words, the amplitude of each order n torque ripple is determined according to the torque output by the motor 14 once the specific configuration of the motor 14 is determined. For this reason, in this embodiment, the basic amplitude calculation unit 31 calculates the first torque command value T １ ＊ and the fundamental amplitude K for each order n ｎ It maintains an amplitude map that has been pre-associated. The basic amplitude calculation unit 31 then refers to this amplitude map to calculate the first torque command value T １ ＊ The corresponding fundamental amplitude K ｎ Perform the calculation.

[0026] The correctable amount calculation unit 32 calculates the first torque command value T １ ＊ and maximum output torque T ｍａｘ Based on this, for torque ripple compensation, the first torque command value T １ ＊By the correction, the magnitude of torque that can be substantially varied (hereinafter referred to as the correction possible amount) is calculated. In the present embodiment, the correction possible amount calculation unit 32 calculates the maximum output torque T ｍａｘ and the deviation of the first torque command value T １ ＊ (hereinafter referred to as torque margin Δ Ｔ ). That is, the correction possible amount calculation unit 32 calculates the torque margin Δ １ ＊ as a parameter representing the correction possible amount of the first torque command value T ｍａｘ by subtracting the first torque command value T １ ＊ from the maximum output torque T Ｔ .

[0027] Note that the maximum output torque T ｍａｘ of the electric motor 14 is determined by the rotational speed ω. In the present embodiment, the maximum output torque T ｍａｘ is known, and it is assumed that the correction possible amount calculation unit 32 can appropriately obtain the maximum output torque T ｍａｘ [[ID=2⑥]]according to the rotational speed ω.

[0028] The filter processing unit 33 performs filtering processing to reduce the natural vibration component of the power train 11 with respect to the rotational speed ω of the electric motor ①. The filter processing unit 33 is configured by a band-stop filter that selectively reduces the natural vibration component of the power train 11 and allows other components to pass through. In the present embodiment, the filter processing unit 33 is particularly configured by a so-called notch filter.

[0029] The priority setting unit 34 sets the priority P ｎ for compensating the torque ripple of each order n according to the rotational speed ω of the electric motor 14. In the present embodiment, the priority setting unit 34 sets the priority P ｎ of each order n according to the filtered rotational speed ω.

[0030] The amplitude correction unit 35 corrects the correction possible amount (torque margin Δ １ ＊ ) of the first torque command value T Ｔ , and the priority P ｎBased on this, the basic amplitude K of each order n ｎ is corrected to calculate the corrected amplitude K ｎ ′ of each order n.

[0031] Specifically, the amplitude correction unit 35 calculates the sum of the basic amplitudes K of each order n (hereinafter referred to as the total basic amplitude K ｎ ). Then, the amplitude correction unit 35 determines whether this total basic amplitude K Σ exceeds the torque margin Δ Σ . That is, the amplitude correction unit 35, in all orders n to be compensated, when calculating the compensation torque (basic compensation torque T Ｔ ) using the basic amplitude K <000009l>, determines whether the corrected second torque command value T Σ ２ [[ID=2u]] ＊ [[ID=Zl]] exceeds the maximum output torque T ｍａｘ . In this embodiment, for the sake of distinction, the compensation torque using the basic amplitude K ｎ in all orders n is referred to as the basic compensation torque T Σ , and the compensation torque calculated using the corrected amplitude K ｎ ′ in at least one order n is simply referred to as the compensation torque T Σ ′.

[0032] When the total basic amplitude K Σ is larger than the torque margin Δ Ｔ , if the first torque command value T ｎ is corrected by the basic compensation torque T Σ using the basic amplitude K １ , the corrected second torque command value T ＊ ２ will exceed the maximum output torque T ＊ . Therefore, in this embodiment, the amplitude correction unit 35 individually corrects the basic amplitude K of each order n according to the priority P ｍａｘ Ｔ so that among the torque ripples of a plurality of orders n, the torque ripples of some orders n are preferentially compensated within the range of the torque margin Δ ｎ . As a result, the corrected amplitude K of each order n ｎ ｎ ​​The ' is set. In other words, the amplitude correction unit 35 limits the compensation of lower-order n torque ripples according to the rotational speed ω, and prioritizes the compensation of higher-order n torque ripples, thereby setting the torque margin Δ Ｔ Within the range, priority P ｎ Accordingly, the correction amplitude K of each order n ｎ Set ′. Specifically, the amplitude correction unit 35 sets priority P ｎ Torque margin Δ in descending order Ｔ It uses the following. And the amplitude correction unit 35 has priority P ｎ Accordingly, torque margin Δ Ｔ Correction amplitude K of order n exceeding ｎ ′ is the fundamental amplitude K ｎ Set it to a value smaller than that.

[0033] For example, priority P ｎ The fundamental amplitude K is ranked 1st and 2nd. ｎ Torque margin Δ Ｔ When it falls within the range, the amplitude correction unit 35 prioritizes P ｎ The corrected amplitude K is the 1st and 2nd place. ｎ ′, those fundamental amplitude K ｎ Set to the same value as above. Then, the amplitude correction unit 35 sets priority P ｎ The corrected amplitude K is in 3rd place. ｎ ′, its fundamental amplitude K ｎ Set to a value smaller than [value]. Also, the amplitude correction unit 35 has priority P ｎ Correction amplitude K for ranks 4 and below ｎ ', their fundamental amplitude K ｎ Regardless of the magnitude, it is set to zero. As a result, the amplitude correction unit 35 sets priority P ｎ The basic amplitude K for 1st to 3rd place ｎ The sum of the torque margin Δ Ｔ It will be contained within. As a result, priority P ｎ Torque ripples of order n with 1st and 2nd priority are preferentially and ideally compensated. Also, priority P ｎ Torque ripples of order n with a third priority are partially but preferentially compensated. And priority P ｎ Torque ripples of order n below 4th rank will no longer be compensated for.

[0034] Note that the total fundamental amplitude K Σ Torque margin Δ Ｔ If the following conditions are met, the fundamental amplitude K ｎ Basic Compensation Torque T using Σ The first torque command value T １ ＊ Even after correction, the corrected second torque command value T ２ ＊ The maximum output torque T ｍａｘ The following applies. Therefore, the amplitude correction unit 35 corrects the amplitude K of each order n. ｎ ′ represents the fundamental amplitude K of each order n. ｎ Set to the same value as: Total fundamental amplitude K Σ Torque margin Δ Ｔ When the following conditions are met, the amplitude correction unit 35 substantially does not perform correction, and the basic amplitude K ｎ The corrected amplitude K remains as is. ｎ It is output as '. In this case, all torque ripples of order n are ideally compensated.

[0035] The fundamental wave calculation unit 36 ​​calculates the first torque command value T. １ ＊ Based on the rotational speed ω and rotational angle θ, the fundamental wave F is determined for each order n of the torque ripple. ｎ The calculation is performed. Fundamental wave F ｎ This represents the temporal change of the torque ripple for each order n. Fundamental wave F ｎ This is the order n, rotational speed ω, time t, rotational angle θ, and the phase φ of the torque ripple for each order n. ｎ Using this, it is expressed by the following equation (1). As can be seen from equation (1), the angular frequency (angular vibration frequency) of the torque ripple is n times (order multiple) the rotational speed ω of the electric motor 14.

[0036]

[0037] Phase φ ｎ This is determined according to the torque output by the electric motor 14 once the specific configuration of the electric motor 14 is determined. For this reason, in this embodiment, the fundamental wave calculation unit 36 ​​determines the first torque command value T １ ＊ and the phase φ of each order n ｎIt maintains a pre-associated phase map. The fundamental wave calculation unit 36 ​​then refers to this phase map to determine the first torque command value T １ ＊ The corresponding phase φ ｎ Perform the calculation.

[0038] Note that the time t in equation (1) is the time after so-called look-ahead compensation (look-ahead compensation time). That is, the time t in equation (1) is the first torque command value T １ ＊ and the second torque command value T ２ ＊ This is a time parameter that compensates for calculation delays, sampling and hold delays, and current control response delays. When the carrier frequency f (not shown) of the PWM control is changed, the fundamental wave calculation unit 36 ​​adjusts the value of time t used in equation (1) accordingly.

[0039] The compensation torque calculation unit 37 calculates the correction amplitude K for each order n. ｎ ' and fundamental wave F ｎ Using this, the compensating torque T Σ The ' is calculated. Specifically, the compensation torque calculation unit 37 calculates the corrected amplitude K ｎ ' and fundamental wave F ｎ By multiplying by this, the compensation value for each order n (hereinafter referred to as each order compensation value T) is obtained. ｎ The ') is calculated. Then, the compensation torque calculation unit 37 calculates each order compensation value T ｎ By summing ', the final compensation torque T Σ ' is calculated. That is, the compensated torque T Σ ' is expressed by the following equation (2).

[0040]

[0041] Figure 4 shows the first torque command value T. １ ＊ Amplitude of torque ripple (fundamental amplitude K) ｎ This graph schematically shows the first torque command value T. As shown in Figure 4, typically, １ ＊ When it is relatively small, the first torque command value T １ ＊ The larger the absolute value of K, the greater the fundamental amplitude.ｎ It increases. However, the first torque command value T １ ＊ When it is relatively large, the first torque command value T １ ＊ The larger the absolute value of K, the greater the fundamental amplitude. ｎ It decreases. The amplitude map held by the basic amplitude calculation unit 31 is the basic amplitude K as described above. ｎ This is a pre-recorded record of the changes.

[0042] Figure 5 shows priority P ｎ This is an explanatory diagram showing the settings. In Figure 5, ω Ａ , ω Ｂ , ω Ｃ , and, ω Ｄ ω (natural angular frequency) is the rotational speed at which the powertrain 11 causes resonance (natural vibration). Here, ω Ａ <ω Ｂ <ω Ｃ <ω Ｄ Therefore, ω Ａ , ω Ｂ , ω Ｃ , and, ω Ｄ This is called the natural angular frequency.

[0043] Also, for simplicity, within the torque ripple of each order n, the natural angular frequency ω Ａ Let "A" be the order n of the torque ripple that causes the strongest vibration, and let P be the priority of the torque ripple of order A. ｎ P Ａ This is expressed as follows: That is, the angular frequency nω is equal to the natural angular frequency ω Ａ The order n of the torque ripple that is closest to and most likely to cause resonance with natural angular frequency ωA is simply represented as "A".

[0044] Other natural angular frequencies ω Ｂ ,ω Ｃ ,ω Ｄ The same applies to the natural angular frequency ω. Ｂ Let B be the order n of the torque ripple that causes the strongest vibration, and P be the priority of the torque ripple of order B. ｎ P Ｂ This is expressed as follows: Natural angular frequency ω ＣLet C be the order n of the torque ripple that causes the strongest vibration, and P be the priority of the torque ripple of order C. ｎ P Ｂ It is expressed as follows. Similarly, the natural angular frequency ω Ｄ Let D be the order n of the torque ripple that causes the strongest vibration, and P be the priority of the torque ripple of order D. ｎ P Ｄ It is represented as follows.

[0045] As shown in Figure 5, priority P of torque ripple of order A Ａ The angular frequency nω (= Aω) is the closest specific natural angular frequency ω Ａ The rotational speed ω is maximized at this natural angular frequency ω Ａ It is set to become smaller the further it is from the target.

[0046] Priority P of torque ripples of order B, C, and D Ｂ , P Ｃ , P Ｃ The same applies to the following: the priority P of the torque ripple of order B. Ｂ The angular frequency nω (= Bω) is the closest natural angular frequency ω Ｂ The rotational speed ω is maximized at this natural angular frequency ω Ｂ The torque ripple priority P is set to decrease as it moves away from the target. Ｃ The angular frequency nω (= Cω) is the closest natural angular frequency ω Ｃ The rotational speed ω is maximized at this natural angular frequency ω Ｃ The torque ripple is set to decrease as it moves away from the target. The priority P of the torque ripple of order D is also set. Ｄ The angular frequency nω (= Dω) is the closest natural angular frequency ω Ｄ The rotational speed ω is maximized at this natural angular frequency ω Ｄ It is set to become smaller the further it is from the target.

[0047] Priority P Ａ , P Ｂ , P Ｃ , P Ｄ As a result of this setting, the rotation speed ω becomes the natural angular frequency ω Ａ Or, if it is in the vicinity of that, priority P Ａ, P Ｂ , P Ｃ , P Ｄ P Ａ >P Ｂ >P Ｃ >P Ｄ Therefore, the priority order for compensating torque ripples of order A, B, C, and D is A > B > C > D. Thus, the rotational speed ω is equal to the natural angular frequency ω. Ａ Or, if it is in the vicinity of that, the total fundamental amplitude K Σ Torque margin Δ Ｔ When it becomes larger than this, the torque ripple of order A is compensated first. Next is the torque margin Δ Ｔ Within this range, torque ripples of orders B, C, and D are compensated preferentially in this order.

[0048] Rotational speed ω is equal to the natural angular frequency ω Ｂ Or, if it is in the vicinity of that, priority P Ａ , P Ｂ , P Ｃ , P Ｄ P Ｂ >P Ａ = P Ｃ >P Ｄ Therefore, the priority order for compensating torque ripples of order A, B, C, and D is B > A = C > D. Thus, the rotational speed ω is equal to the natural angular frequency ω. Ｂ Or, if it is in the vicinity of that, the total fundamental amplitude K Σ Torque margin Δ Ｔ When it becomes larger than Δ, the torque ripple of order B is compensated first, followed by the torque margin Δ. Ｔ Within this range, torque ripples of order A or order C are preferentially compensated.

[0049] Note that the rotational speed ω is equal to the natural angular frequency ω ＢWhen the torque ripple is either near or near that order, both torque ripples of order A and C have a priority of 2. Therefore, in the relationship between torque ripple of order A and torque ripple of order C, it is arbitrary which torque ripple is given priority in compensation. In this embodiment, for simplicity, it is assumed that torque ripples of orders with relatively smaller corresponding natural angular frequencies are given priority in compensation. This is also true for other rotational speeds ω.

[0050] Rotational speed ω is equal to the natural angular frequency ω Ｃ Or, if it is in the vicinity of that, priority P Ａ , P Ｂ , P Ｃ , P Ｄ P Ｃ >P Ｂ = P Ｄ >P Ａ Therefore, the priority order for compensating torque ripples of order A, B, C, and D is C > B = D > A. Thus, the rotational speed ω is equal to the natural angular frequency ω. Ｃ Or, if it is in the vicinity of that, the total fundamental amplitude K Σ Torque margin Δ Ｔ When it becomes larger than this, the torque ripple of order C is compensated first, followed by the torque margin Δ. Ｔ Within this range, torque ripples of order B or order D are preferentially compensated.

[0051] And the rotational speed ω is equal to the natural angular frequency ω Ａ Or, if it is in the vicinity of that, priority P Ａ , P Ｂ , P Ｃ , P Ｄ P Ｄ >P Ｃ >P Ｂ >P Ａ Therefore, the priority order for compensating torque ripples of order A, B, C, and D is D > C > B > A. Thus, the rotational speed ω is equal to the natural angular frequency ω. Ｄ Or, if it is in the vicinity of that, the total fundamental amplitude K Σ Torque margin Δ Ｔ When it becomes larger than this, the torque ripple of order D is compensated first, followed by the torque margin Δ.Ｔ Within this range, torque ripples of order C, B, and A are compensated preferentially in this order.

[0052] Typically, the higher the order n of the torque ripple, the smaller the natural angular frequency at which resonance occurs. For example, ω Ａ <ω Ｂ <ω Ｃ <ω Ｄ The values ​​of the torque ripple orders A, B, C, and D that produce resonance at each natural angular frequency are A > B > C > D. That is, the smallest natural angular frequency ω Ａ The torque ripple order A that produces the strongest resonance is the largest among orders A, B, C, and D. Also, the largest natural angular frequency ω Ｄ The order D of the torque ripple that produces the strongest resonance is the smallest among orders A, B, C, and D. Therefore, the priority P above... ｎ In the settings, priority P ｎ In effect, the system is configured such that higher-order torque ripples (n) are preferentially compensated for as the rotational speed ω decreases.

[0053] Figure 6 shows the torque command value (T １ ＊ The phase of the torque ripple relative to (phase φ of each order n) ｎ This graph schematically shows the first torque command value T. As shown in Figure 6, typically, １ ＊ This is positive, and the first torque command value T １ ＊ When the absolute value of is relatively small, the first torque command value T １ ＊ The larger the absolute value of φ, the greater the phase φ ｎ It decreases. And the first torque command value T １ ＊ As the absolute value of becomes even larger, the phase φ ｎ It increases and asymptotically approaches a predetermined value. Meanwhile, the first torque command value T １ ＊ If it is negative, the first torque command value T １ ＊ When the absolute value of is relatively small, the first torque command value T １ ＊ The smaller the phase φ becomes, ｎIt increases. And the first torque command value T １ ＊ As the absolute value of becomes even larger, the phase φ ｎ It decreases and asymptotically approaches a predetermined value. The phase map held by the fundamental wave calculation unit 36 ​​is the phase φ as described above. ｎ This is a pre-recorded record of the changes.

[0054] Figure 7 is a flowchart relating to torque ripple compensation. As shown in Figure 7, in step S11, the fundamental wave calculation unit 36 ​​calculates the first torque command value T １ ＊ Based on the rotational speed ω and the rotational angle θ, the fundamental wave F of each order n is determined. ｎ The first torque command value T is calculated. In step S12, the basic amplitude calculation unit 31 calculates the first torque command value T １ ＊ Based on this, the fundamental amplitude K for each order n ｎ The first torque command value T is calculated. In step S13, the correctable amount calculation unit 32 calculates the first torque command value T １ ＊ and maximum output torque T ｍａｘ Based on this, the first torque command value T １ ＊ Torque margin Δ represents the effective amount of correctable torque. Ｔ The following is calculated. In step S14, the priority setting unit 34 sets the priority P for the compensation of each order n of torque ripple according to the rotational speed ω. ｎ Set it.

[0055] In step S15, the amplitude correction unit 35 adjusts the torque margin Δ Ｔ Priority P ｎ Based on this, the fundamental amplitude K for each order n ｎ By correcting each of these, the correction amplitude K of each order n is obtained. ｎ ' is calculated, that is, the torque margin Δ Ｔ Within the range, priority P is set such that some of the torque ripples of order n among multiple torque ripples of order n are preferentially compensated. ｎ Accordingly, the fundamental amplitude K for each order n ｎ These are set individually. Note that priority P ｎ Since it is determined according to the rotational speed ω, as above, priority P ｎThe fundamental amplitude K based on this ｎ Correction (correction amplitude K) ｎ The calculation of ' is a correction (calculation) that corresponds to the rotation speed ω.

[0056] In step S16, the compensation torque calculation unit 37 calculates the correction amplitude K for each order n. ｎ ' and fundamental wave F ｎ Compensation torque T using Σ The ' is calculated. In step S17, the torque command value correction unit 22 calculates the compensated torque T Σ Using ′, the first torque command value T １ ＊ By correcting this, the second torque command value T ２ ＊ This second torque command value T is calculated. ２ ＊ In the current control unit 23, the phase current i ＵＶＷ It is used to calculate the target value for the phase current i. ＵＶＷ It is controlled to match this target value.

[0057] This results in the total fundamental amplitude K Σ Torque margin Δ Ｔ Larger than, fundamental amplitude K ｎ Basic Compensation Torque T using Σ The first torque command value T １ ＊ After correction, the corrected second torque command value T ２ ＊ Maximum output torque T ｍａｘ When the torque ripple exceeds a certain value, some of the torque ripples of order n among the multiple order n torque ripples are preferentially (selectively) compensated, depending on the rotational speed ω of the electric motor 14. As a result, the torque ripple is always compensated more appropriately than when all order n torque ripples that are subject to compensation are compensated equally.

[0058] Figure 8 shows the corrected amplitude K. ｎ This is a flowchart related to the calculation of '. For the sake of explanation, we assume that the three order n torque ripples are the ones to be compensated. Also, here, the fundamental amplitude K ｎ and corrected amplitude K ｎ ' is priority P ｎThe order (priority) determined by the following is indicated. Specifically, the fundamental amplitude K of order n with priority 1 is shown. ｎ and corrected amplitude K ｎ ' represents K ｎ [1] and K ｎ '[1]. The fundamental amplitude K of order n with priority 2. ｎ and corrected amplitude K ｎ ' represents K ｎ [2] and K ｎ '[2]. Similarly, the fundamental amplitude K of order n with priority 3. ｎ and corrected amplitude K ｎ ' represents K ｎ [3] and K ｎ '[3].

[0059] As shown in Figure 8, in step S20, the amplitude correction unit 35 adjusts the total fundamental amplitude K Σ = K ｎ [1] + K ｎ [2] + K ｎ [3] is calculated and this becomes the torque margin Δ Ｔ This is compared with the fundamental amplitude K. ｎ Basic Compensation Torque T using Σ The first torque command value T １ ＊ When correcting, the corrected second torque command value T ２ ＊ Maximum output torque T ｍａｘ Determine whether it exceeds a certain value.

[0060] In step S20, the total fundamental amplitude K Σ Torque margin Δ Ｔ Below (K Σ ≤Δ Ｔ ) and the fundamental amplitude K ｎ Basic Compensation Torque T using Σ The first torque command value T １ ＊ Even when correcting, the corrected second torque command value T ２ ＊ Maximum output torque T ｍａｘ When it is determined that it does not exceed the threshold, the amplitude correction unit 35 substantially adjusts the fundamental amplitude K of each order n. ｎWithout correction, the fundamental amplitude K ｎ The corrected amplitude K remains as is. ｎ It outputs as '. That is, the amplitude correction unit 35 is K ｎ '[1] = K ｎ [1], K ｎ '[2] = K ｎ [2], K ｎ '[3] = K ｎ Set to [3].

[0061] On the other hand, in step S20, the total fundamental amplitude K Σ Torque margin Δ Ｔ Larger than (K Σ >Δ Ｔ ), basic amplitude K ｎ Basic Compensation Torque T using Σ The first torque command value T １ ＊ When correcting, the corrected second torque command value T ２ ＊ Maximum output torque T ｍａｘ If it is determined that the value exceeds a certain limit, the process proceeds to step S21.

[0062] In step S21, the amplitude correction unit 35 adjusts the fundamental amplitude K of order n, which has priority 1. ｎ [1] Torque margin Δ Ｔ Compare it with the fundamental amplitude K. ｎ [1] is torque margin Δ Ｔ When it is greater than (K ｎ [1] > Δ Ｔ ), proceed to step S22, where the amplitude correction unit 35 corrects the first-priority order n correction amplitude K ｎ '[1] Torque margin Δ Ｔ Set to (K ｎ ′[1] = Δ Ｔ ). Furthermore, the amplitude correction unit 35 corrects the amplitude K of the order n of priority 2nd and 3rd. ｎ '[2], K ｎ Set [3] to zero. That is, K ｎ ′[1] = Δ Ｔ , K ｎ '[2] = 0, K ｎ '[3] = 0. This results in a torque margin Δ ＴWithin this range, torque ripples of order n with priority 1 are preferentially compensated. However, torque ripples of order n with priority 2 and 3 are no longer compensated.

[0063] In step S21, the fundamental amplitude K ｎ [1] is torque margin Δ Ｔ Below (K ｎ [1] ≤ Δ Ｔ If this is the case, the process proceeds to step S23, and the amplitude correction unit 35 corrects the first-order n correction amplitude K ｎ ′[1] with fundamental amplitude K ｎ Set to the same value as in [1]. Then, in step S24, the amplitude correction unit 35 sets the fundamental amplitude K of order n, which has a priority of 2. ｎ [2] Torque margin Δ Ｔ From the fundamental amplitude K ｎ [1] Subtracted residual (Δ Ｔ -K ｎ Compare with [1]).

[0064] In step S24, the fundamental amplitude K ｎ [2] is the residual (Δ Ｔ -K ｎ [1]) If it is greater than the above, the process proceeds to step S25, and the amplitude correction unit 35 corrects the correction amplitude K of order n, which has a priority of 2. ｎ '[2] is this residual (Δ Ｔ -K ｎ Set to the value of [1]). The amplitude correction unit 35 also sets the corrected amplitude K of order n, which has a priority of 3rd place. ｎ Set [3] to zero. That is, K ｎ '[1] = K ｎ [1], K ｎ '[2] = Δ Ｔ -K ｎ [1], K ｎ '[3] = 0. As a result, the torque ripple of order n with priority 1 is ideally compensated, and the torque ripple of order n with priority 2 is compensated by the torque margin Δ Ｔ It will be partially compensated within that range. And, torque ripple with priority 3 will no longer be compensated.

[0065] In step S24, the fundamental amplitude K ｎ [2] is the residual (Δ Ｔ-K ｎ [1]) If the condition is less than or equal to the above, the process proceeds to step S26, and the amplitude correction unit 35 corrects the amplitude K of order n, which has a priority of 2. ｎ '[2] with fundamental amplitude K ｎ Set to the same value as in [2]. Then, in step S27, the amplitude correction unit 35 sets the fundamental amplitude K of the third priority order. ｎ [3] Torque margin Δ Ｔ From the fundamental amplitude K ｎ [1], K ｎ [2] Subtracted residual (Δ Ｔ - (K ｎ [1] + K ｎ Compare with [2])).

[0066] In step S27, the fundamental amplitude K ｎ [3] is the residual (ΔT - (K ｎ [1] + K ｎ [2])) If it is greater than the above, the process proceeds to step S28, and the amplitude correction unit 35 corrects the correction amplitude K of order n, which has priority 3. ｎ '[3] is this residual (Δ Ｔ - (K ｎ [1] + K ｎ Set to the value of [2])). That is, K ｎ '[1] = K ｎ [1], K ｎ '[2] = K ｎ [2], K ｎ '[3] = Δ Ｔ - (K ｎ [1] + K ｎ [2]) This results in ideally compensating for the order n torque ripples of priority 1 and 2, and the torque ripple of order n torque 3 being compensated for by the torque margin Δ Ｔ It will be partially compensated within the limits of the scope.

[0067] In step S27, the fundamental amplitude K ｎ [3] is the residual (Δ Ｔ - (K ｎ [1] + K ｎ [2])) If the condition is less than or equal to the condition, the process proceeds to step S29, and the amplitude correction unit 35 adjusts the correction amplitude K of order n, which has priority 3. ｎ '[3] with fundamental amplitude K ｎSet to the same value as in [3]. That is, K ｎ '[1] = K ｎ [1], K ｎ '[2] = K ｎ [2], K ｎ '[3] = K ｎ [3]

[0068] Thus, the amplitude correction unit 35 adjusts the fundamental amplitude K of each order n. ｎ The sum of (total fundamental amplitude K) Σ ) is the correctable amount (Δ Ｔ When it exceeds Δ, the correctable amount ( Ｔ Within the range of ), the torque ripple of order n with the highest priority is preferentially compensated, and the correction amplitude K for each order n is set accordingly. ｎ Determine the size of ′.

[0069] In particular, the amplitude correction unit 35 has priority P ｎ Correction amount (Δ) Ｔ ) is used. That is, the amplitude correction unit 35 uses the fundamental amplitude K of order n, which has a high priority for compensation. ｎ Torque margin Δ Ｔ Prioritizes the use of P. The amplitude correction unit 35 then prioritizes P. ｎ Correctable amount (Δ Ｔ Correction amplitude K of order n exceeding ) ｎ ′ is the fundamental amplitude K ｎ Make it smaller than this. This will result in priority P ｎ Torque ripples of higher order n are preferentially compensated.

[0070] Note that here, the corrected amplitude K ｎ To explain the basic concept of the calculation of ', a flow including steps S20, S27, and S29 was conveniently described, but in step S20 K Σ >Δ Ｔ In this case, the process does not effectively proceed from step S27 to step S29. Therefore, when step S20 is executed, steps S27 and S29 can be omitted. Also, when steps S27 and S29 are executed, step S20 can be omitted.

[0071] Furthermore, although it is assumed here that three order n torque ripples are subject to compensation, the amplitude correction unit 35 also corrects the amplitude K in the same way as above when two order n torque ripples are subject to compensation, and when four or more order n torque ripples are subject to compensation. ｎ Perform the calculation '.

[0072] The following describes the operation of torque ripple compensation according to this embodiment, in comparison with the comparative example. The comparative example is priority P. ｎ No settings are made, and the fundamental amplitude K of all orders n that are subject to compensation is not set. ｎ Using the same as the basic compensatory torque T Σ The calculation is performed. Then, the total fundamental amplitude K Σ Torque margin Δ Ｔ If it exceeds (correctable amount), in the comparative example, the basic compensation torque T Σ Multiply this by a gain g (g < 1). This gives the basic compensated torque T Σ This is reduced, and the corrected second torque command value T ２ ＊ Maximum output torque T ｍａｘ It is included below.

[0073] Figure 9 shows the compensation torque T in the comparative example. Σ This is an explanatory diagram showing the basic compensation torque T. Σ The breakdown is as follows: the fundamental amplitude K of each order n ｎ This is an explanatory diagram. Here, four order torque ripples A, B, C, and D are subject to compensation. Figure 9(B) shows the compensated torque T reduced by the gain g. Σ ″=gT Σ This is an explanatory diagram showing the breakdown of those components.

[0074] As shown in Figure 9(A), the total fundamental amplitude K Σ = K Ａ +K Ｂ +K Ｃ +K Ｄ is torque margin Δ Ｔ It is larger than that. Therefore, in the comparative example, the fundamental amplitude K Ａ , K Ｂ , K Ｃ , K Ｄ The basic compensation torque T calculated using ΣBy multiplying by the gain g, the final compensated torque T is obtained as shown in Figure 9(B). Σ The size of the " is the torque margin Δ Ｔ It is reduced to this extent. As a result, as shown in Figure 9(B), in the comparative example, the compensation torque T Σ In this case, the amplitudes of each order A, B, C, and D are reduced substantially uniformly by the gain g. For example, the compensated torque T Σ The amplitude of the torque ripple of order A contained in '' is K Ａ ``=gK Ａ <K Ａ It is reduced to the same extent. Similarly, the compensating torque T Σ The amplitude of the torque ripple of order B contained in '' is K Ｂ ``=gK Ｂ <K Ｂ As a result, the compensating torque T Σ The amplitude of the torque ripple of order C contained in '' is K Ｃ ``=gK Ｃ <K Ｃ This is the result. Also, the compensating torque T Σ The amplitude of the torque ripple of order D contained in '' is K Ｄ ``=gK Ｄ <K Ｄ Therefore, in the comparative example, the compensation for torque ripple of all orders A, B, C, and D is partial.

[0075] Figure 10 shows the compensation torque T of this embodiment. Σ This is an explanatory diagram showing ′. Figure 10(A) is similar to Figure 9(A) in that it shows the basic compensating torque T Σ The breakdown is as follows: the fundamental amplitude K of each order n ｎ This is an explanatory diagram. Here, as in the comparative example, four order torque ripples A, B, C, and D are assumed to be compensated. Furthermore, here, priority P is based on rotational speed ω. ｎ The priority order for compensation is assumed to be A > B > C > D. Figure 10(B) shows the corrected amplitude K. ｎ Compensation torque T calculated using ' Σ The breakdown of ' is the correction amplitude K for each order n. ｎ This is an explanatory diagram shown by ′.

[0076] As shown in Figure 10(A), the total fundamental amplitude K is shown here. Σ = KＡ +K Ｂ +K Ｃ +K Ｄ is torque margin Δ Ｔ It is larger than that. Therefore, in this embodiment, the fundamental amplitude K Ａ , K Ｂ , K Ｃ , K Ｄ By individually correcting these, the final compensated torque T is obtained as shown in Figure 10(B). Σ The size of ′ is the torque margin Δ Ｔ It is reduced to this extent. In this embodiment, the torque margin Δ Ｔ Since they are used in order of highest priority, the correction amplitude K Ａ ', K Ｂ ', K Ｃ ', K Ｄ ' represents K Ａ ' = K Ａ , K Ｂ ' = K Ｂ , K Ｃ ′=Δ Ｔ - (K Ａ +K Ｂ ) <K Ｃ , K Ｄ It is set to = 0. Therefore, in this embodiment, torque ripple of order A and order B is ideally compensated, and torque ripple of order C is partially compensated. Also, in the scene shown here, torque ripple of order D is not compensated.

[0077] Comparing this embodiment with the comparative example, the compensation torque T of this embodiment is Σ ' and the compensation torque T of the comparative example Σ The size of the " is the torque margin Δ Ｔ Therefore, the total fundamental amplitude K Σ Torque margin Δ Ｔ If it is greater than, in both this embodiment and the comparative example, the corrected second torque command value T ２ ＊ The maximum output torque T ｍａｘ It fits within this. However, the compensating torque T of this embodiment Σ ' and the compensation torque T of the comparative example Σ "However, as mentioned above, the breakdown is different. Due to this difference in the breakdown, the total fundamental amplitude K ΣTorque margin Δ Ｔ When the value is greater than the comparative example, torque ripple is compensated better in this embodiment than in the comparative example.

[0078] Figure 11 is a graph showing the effect of the comparative example. Here, for ease of illustration, the torque ripple T of three orders A, B, and C is shown. Ａ , T Ｂ , T Ｃ The following is considered to be covered by compensation: Torque ripple T of order A Ａ Among these, the natural angular frequency ω Ａ This is the torque ripple that produces the strongest resonance in [location]. Torque ripple T of order B Ｂ is the natural angular frequency ω Ｂ (>ω Ａ This is the torque ripple that produces the strongest resonance. And then there is the torque ripple of order C T Ｃ is the natural angular frequency ω Ｃ (>ω Ｂ This is the torque ripple that produces the strongest resonance. The horizontal axis of each graph in Figures 11(A) to 11(E) is rotational speed ω.

[0079] Figure 11(A) shows the fundamental amplitude K Ａ Torque ripple T of order A having Ａ The dotted line shows the torque ripple of order A, where the amplitude is reduced by the gain g. Ａ The value is shown by a solid line. Figure 11(B) shows the fundamental amplitude K. Ｂ Torque ripple T of order B having Ｂ The dotted line shows the torque ripple of order B, with the amplitude reduced by the gain g. Ｂ The value is shown by a solid line. Figure 11(C) shows the fundamental amplitude K. Ｃ Torque ripple T of order C having Ｃ The dotted line shows the torque ripple of order C, whose amplitude is reduced by the gain g. Ｃ This is shown by a solid line. Figure 11(D) shows the basic compensating torque T Σ The dotted line indicates the compensation torque T of the comparative example. Σ The value is shown as a solid line. Figure 11(E) shows the variation in rotational speed ω that occurs in the comparative example, shown as a solid line. The dashed line in Figure 11(E) shows the change in rotational speed ω in the ideal case.

[0080] Total fundamental amplitude KΣ Torque margin Δ Ｔ If it exceeds this, in the comparative example, as shown in Figure 11(D), the basic compensation torque T Σ This is the corrected second torque command value T. ２ ＊ Maximum output torque T ｍａｘ The gain g compensates the torque T so that it fits within the range. Σ The gain g is reduced to ". The gain g does not depend on the rotational speed ω of the electric motor 14, for example, the total fundamental amplitude K Σ and torque margin Δ Ｔ It is determined by the ratio of ω. Therefore, the natural angular frequency ω Ａ Frequency band Z centered on Ａ , natural angular frequency ω Ｂ Frequency band Z centered on Ｂ , and the natural angular frequency ω Ｃ Frequency band Z centered on Ｃ In either case, the compensation torque T of the comparative example Σ " is the basic compensating torque T, regardless of the rotational speed ω. Σ The amplitude is reduced to a constant value. Therefore, the compensation torque T of the comparative example Σ The torque ripple compensated by " is a torque ripple T whose amplitude is reduced to a constant level regardless of the rotational speed ω, as shown in Figures 11(A) to 11(C). Ａ ``, T Ｂ ``, T Ｃ It becomes ".

[0081] As a result, in the comparative example, the first torque command value T １ ＊ Maximum output torque T ｍａｘ It approaches the total fundamental amplitude K Σ Torque margin Δ Ｔ When the situation exceeds this, as shown in Figure 11(E), vibration occurs in the rotational speed ω of the electric motor 14. This vibration in rotational speed ω then causes torque ripple. In other words, in the comparative example, the total fundamental amplitude K Σ Torque margin Δ Ｔ If the situation exceeds this limit, the torque ripple will not be sufficiently reduced.

[0082] Figure 12 is a graph illustrating the operation of this embodiment. Here, as with Figure 11, for ease of illustration, three torque ripples of order A, B, and C are shown. Ａ , T Ｂ , T Ｃ This is covered by the compensation. Also, the horizontal axis of each graph in Figures 12(A) to 12(F) is the rotational speed ω.

[0083] Figure 12(A) shows the priority levels P of each order A, B, and C according to the rotational speed ω. ｎ The settings are shown. Figure 12(B) shows the fundamental amplitude K Ａ Torque ripple T of order A having Ａ This is shown by a dotted line, and the corrected amplitude K Ａ Torque ripple T of order A having ′ Ａ The ' is shown by a solid line. Figure 12(C) shows the fundamental amplitude K Ｂ Torque ripple T of order B having Ｂ This is shown by a dotted line, and the corrected amplitude K Ｂ Torque ripple T of order B having ′ Ｂ The ' is shown by a solid line. Figure 12(D) shows the fundamental amplitude K. Ｃ Torque ripple T of order C having Ｃ This is shown by a dotted line, and the corrected amplitude K Ｃ Torque ripple T of order C having ' Ｃ This is shown by a solid line. Figure 12(E) shows the basic compensating torque T Σ The dotted line indicates the compensation torque T of this embodiment. Σ The ' is shown as a solid line. Figure 12(F) shows the variation in rotational speed ω that occurs in this embodiment as a solid line. In Figure 12(E), the dashed line shows the change in ideal rotational speed ω, and the dotted line shows the variation in rotational speed ω that occurs in the comparative example.

[0084] As shown in Figure 12(A), in this embodiment, the torque ripple T of each order A, B, and C is determined according to the rotational speed ω of the electric motor 14. Ａ , T Ｂ , T Ｃ Regarding compensation, priority P ｎ This will be set.

[0085] natural angular frequency ω Ａ Frequency band Z centered on Ａ So, what is the priority P for each degree A, B, and C? ｎ PＡ >P Ｂ >P Ｃ That is, frequency band Z Ａ The priority order for compensation in this case is Order A > Order B > Order C. Therefore, when the rotational speed ω of the motor 14 is in frequency band Z Ａ If it belongs to this category, as shown in Figure 12(B), the torque ripple T of order A Ａ ′ Amplitude (corrected amplitude K Ａ ′) is, for example, the fundamental amplitude K Ａ It is set to be equal to . Also, as shown in Figure 12(C), the torque ripple T of order B Ｂ ′ Amplitude (corrected amplitude K Ｂ ′) is, for example, the torque margin Δ Ｔ The basic amplitude K depends on the remaining amount. Ｂ This is reduced. And, as shown in Figure 12(D), the torque ripple T of order C Ｃ ′ Amplitude (corrected amplitude K Ｃ ′) is set to zero, for example.

[0086] natural angular frequency ω Ｂ Frequency band Z centered on Ｂ Of these, the lower frequency band Z Ｂ１ So, what is the priority P for each degree A, B, and C? ｎ P Ｂ >P Ａ >P Ｃ That is, frequency band Z Ｂ１ The priority order for compensation in this case is Order B > Order A > Order C. Therefore, when the rotational speed ω of the motor 14 is in frequency band Z Ｂ１ If it belongs to this category, as shown in Figure 12(C), the torque ripple T of order B Ｂ ′ Amplitude (corrected amplitude K Ｂ ′) is, for example, the fundamental amplitude K Ａ It is set to be equal to . Also, as shown in Figure 12(B), the torque ripple T of order A Ａ ′ Amplitude (corrected amplitude K Ａ ′) is, for example, the fundamental amplitude K depending on the remaining amount of torque margin ΔT. Ａ This is reduced. And, as shown in Figure 12(D), the torque ripple T of order C Ｃ ′ Amplitude (corrected amplitude K Ｃ′) is set to zero, for example.

[0087] natural angular frequency ω Ｂ Frequency band Z centered on Ｂ Of these, the higher frequency band Z Ｂ２ So, what is the priority P for each degree A, B, and C? ｎ P Ｂ >P Ｃ >P Ａ That is, frequency band Z Ｂ２ The priority order for compensation in this case is Order B > Order C > Order A. Therefore, when the rotational speed ω of the motor 14 is in frequency band Z Ｂ２ If it belongs to this category, as shown in Figure 12(C), the torque ripple T of order B Ｂ ′ Amplitude (corrected amplitude K Ｂ ′) is, for example, the fundamental amplitude K Ａ It is set to be equal to . Also, as shown in Figure 12(D), the torque ripple T of order C Ｃ ′ Amplitude (corrected amplitude K Ｃ ′) is, for example, the torque margin Δ Ｔ The basic amplitude K depends on the remaining amount. Ｃ This is reduced. And, as shown in Figure 12(B), the torque ripple T of order A Ａ ′ Amplitude (corrected amplitude K Ａ ′) is set to zero, for example.

[0088] natural angular frequency ω Ｃ Frequency band Z centered on Ｃ So, what is the priority P for each degree A, B, and C? ｎ P Ｃ >P Ｂ >P Ａ That is, frequency band Z Ｃ The priority order for compensation in this case is Order C > Order B > Order A. Therefore, when the rotational speed ω of the motor 14 is in frequency band Z Ｃ If it belongs to this category, as shown in Figure 12(D), the torque ripple T of order C Ｃ ′ Amplitude (corrected amplitude K Ｃ ′) is, for example, the fundamental amplitude K Ｃ It is set to be equal to . Also, as shown in Figure 12(C), the torque ripple T of order B Ｂ ′ Amplitude (corrected amplitude K Ｂ′) is, for example, the torque margin Δ Ｔ The basic amplitude K depends on the remaining amount. Ｂ This is reduced. And, as shown in Figure 12(A), the torque ripple T of order A Ａ ′ Amplitude (corrected amplitude K Ａ ′) is set to zero, for example.

[0089] At this time, the above torque ripple T Ａ ', T Ｂ ', T Ｃ The total of ' is the compensation torque T Σ ′ is as shown in Figure 12(E), and the basic compensated torque T Σ The amplitude is reduced. Therefore, the corrected second torque command value T ２ ＊ The maximum output torque T ｍａｘ It will fit within the following limits. However, as mentioned above, the compensating torque T Σ ' is Torque Ripple T Ａ , T Ｂ , T Ｃ It does not contain them equally, but rather the frequency band Z to which the rotational speed ω of the electric motor 14 belongs. Ａ , Z Ｂ１ , Z Ｂ２ , Z Ｃ Breakdown according to (torque ripple T Ａ , T Ｂ , T Ｃ The component ratio of changes. Therefore, the compensating torque T Σ ' is simply the basic compensated torque T Σ This does not reduce the amplitude, but rather the basic compensated torque T Σ The waveform also changes accordingly. Specifically, the compensated torque T Σ ' represents the corrected second torque command value T ２ ＊ The maximum output torque T ｍａｘ The system is configured to prioritize compensating for torque ripples of a certain order that cause resonance in the powertrain 11, while keeping the following conditions in mind.

[0090] As a result, in this embodiment, the first torque command value T １ ＊ Maximum output torque T ｍａｘ It approaches the total fundamental amplitude K Σ Torque margin ΔＴ Even when the situation exceeds this, as shown in Figure 12(F), the vibration of the rotational speed ω is suppressed more than in the comparative example (dotted line). In other words, in this embodiment, the total fundamental amplitude K Σ Torque margin Δ Ｔ Even in situations exceeding this limit, torque ripple is effectively reduced.

[0091] In the above embodiment, the torque ripple calculation unit 21 has priority P ｎ Although the settings are configured, the explicit priority P ｎ The calculation (setting) of can be omitted. That is, the torque ripple calculation unit 21 simply corrects the amplitude K according to the rotational speed ω, etc. ｎ The ' can be calculated. For example, the torque ripple calculation unit 21 calculates the fundamental amplitude K ｎ Torque margin Δ Ｔ (Correctable amount), and rotational speed ω and correction amplitude K ｎ Refer to the corrected amplitude torque map (not shown) that corresponds to ' and, corrected amplitude K ｎ The amplitude correction unit 35 can be configured to calculate '. In this case, the corrected amplitude map is predetermined based on experiments or simulations, etc., to preferentially compensate for some order n torque ripples according to the rotational speed ω, similar to the embodiment described above.

[0092] Furthermore, although the above embodiment described the control of an electric motor 14 used in the powertrain 11 of an electric vehicle 100, it is not limited to this. The above control (torque ripple compensation) for the electric motor 14 is also suitable for the control of other electric motors.

[0093] As described above, the control method for the electric motor according to the above embodiment is based on the torque command value (T １ ＊ This is a control method for the electric motor 14 that compensates for torque ripple by correcting the fundamental amplitude K of the torque ripple for each of the multiple orders n. ｎ The fundamental amplitude K of each order n is calculated. ｎ By correcting each of these according to the rotational speed ω of the electric motor 14, the correction amplitude K of each order n is obtained. ｎ ' is calculated, and the corrected amplitude K of each order n is calculated.ｎ Compensation torque T using ' Σ ' is calculated, and this compensation torque T Σ Using ′, the torque command value (T １ ＊ Correct the following:

[0094] Torque command value (T １ ＊ ) is the maximum output torque T ｍａｘ It has a large value close to the fundamental amplitude K ｎ The basic compensated torque T calculated using Σ Torque command value (T １ ＊ When the corrected torque command value (T) is obtained, ２ ＊ ) is the maximum output torque T ｍａｘ If it exceeds this value, the torque ripple is not adequately compensated. However, in the motor control method according to the above embodiment, as described above, the torque command value (T １ ＊ Compensation torque T used for correction of ) Σ ' is the corrected amplitude K, which is corrected for each order n according to the rotational speed ω of the electric motor 14. ｎ It is calculated using '. Therefore, according to the motor control method of the above embodiment, the basic compensatory torque T Σ Using this method, the corrected torque command value (T ２ ＊ ) is the maximum output torque T ｍａｘ Even in situations where it exceeds a certain limit, torque ripple can be effectively reduced. In particular, the basic compensated torque T Σ By multiplying this by a gain g and uniformly limiting the compensation for torque ripple of each order n, the corrected torque command value (T ２ ＊ ) to maximum output torque T ｍａｘ Compared to the case described below (comparative example), the motor control method according to the above embodiment significantly reduces torque ripple.

[0095] In the motor control method according to the above embodiment, the maximum output torque T of the motor 14 ｍａｘ And, the torque command value (T １ ＊ ) and the correctable amount (Δ) which is the deviation of )Ｔ ) is calculated, and priority P is used for compensating for each order n of torque ripple according to the rotational speed ω. ｎ Set the fundamental amplitude K for each order n. ｎ The sum of these is the correctable quantity (Δ Ｔ When it exceeds Δ, the correctable amount ( Ｔ Within the range of priority P ｎ The correction amplitude K for each order n is set so that torque ripples of higher order n are preferentially compensated. ｎ Determine the size of ′.

[0096] Thus, priority P is used for compensating for torque ripple of each order n. ｎ Set the torque margin Δ Ｔ Within the range of (correctable amount), priority P ｎ The correction amplitude K for each order n is set so that torque ripples of higher order n are preferentially compensated. ｎ Once the magnitude of ' is determined, the basic compensatory torque T Σ Using this method, the corrected torque command value (T ２ ＊ ) is the maximum output torque T ｍａｘ Even in situations where the torque exceeds a certain limit, it is particularly effective at reducing torque ripple.

[0097] In the electric motor control method according to the above embodiment, priority P ｎ Correction amount (Δ) Ｔ Use ) and priority P ｎ Correctable amount (Δ Ｔ Correction amplitude K of order n exceeding ) ｎ ′ is the fundamental amplitude K ｎ Make it smaller than that.

[0098] Thus, the fundamental amplitude K of order n has a high priority for compensation. ｎ Torque margin Δ Ｔ If this is used preferentially, some order n torque ripples that should be preferentially compensated according to the rotational speed ω will have a sufficient torque margin Δ Ｔ It is ideally compensated, just as if there were a torque margin Δ. Also, the amplitude is torque margin Δ Ｔ Within the range that fits within, priority P ｎ Accordingly, torque margin Δ ＴCorrection amplitude K of order n exceeding ｎ ′ is the fundamental amplitude K ｎ If it is made smaller than this, the torque ripple of order n will be compensated at least partially. Therefore, if we try to compensate all order n torque ripples that are subject to compensation equally, the corrected torque command value (T ２ ＊ ) is the maximum output torque T ｍａｘ Even in situations where the torque exceeds a certain limit, it is particularly effective at reducing torque ripple.

[0099] In the motor control method according to the above embodiment, the priority P of each order n is ｎ This is the natural angular frequency (e.g., ω) that is closest to the angular frequency (nω) of each order n. Ａ ) is maximized at the natural angular frequency (e.g., ω Ａ It is set to become smaller the further it is from ().

[0100] Thus, the angular frequency nω is the closest specific natural angular frequency (ω Ａ The rotational speed ω is maximized at (etc.), and its natural angular frequency (ω Ａ Priority P ｎ By setting this, the torque ripple of order n that should be preferentially compensated for according to the rotational speed ω of the electric motor 14 is selected particularly appropriately. Therefore, the basic compensated torque T Σ Using this method, the corrected torque command value (T ２ ＊ ) is the maximum output torque T ｍａｘ Even in situations where the torque exceeds a certain limit, it is particularly effective at reducing torque ripple.

[0101] In the electric motor control method according to the above embodiment, priority P ｎ The settings are configured such that higher-order torque ripples (n) are preferentially compensated as the rotational speed ω decreases.

[0102] As mentioned above, typically, the higher the order n of the torque ripple, the smaller the natural angular frequency at which resonance occurs. Therefore, as described above, priority P is set so that torque ripples of higher order n are preferentially compensated as the rotational speed ω decreases. ｎBy setting this, the torque ripple of order n that should be preferentially compensated for according to the rotational speed ω of the electric motor 14 is selected particularly appropriately. Therefore, the basic compensated torque T Σ Using this method, the corrected torque command value (T ２ ＊ ) is the maximum output torque T ｍａｘ Even in situations where the torque exceeds a certain limit, it is particularly effective at reducing torque ripple.

[0103] In the motor control method according to the above embodiment, a filter (e.g., a notch filter) is used to reduce the natural vibration component from the rotational speed ω, and the rotational speed ω with the reduced natural vibration component is used to control priority P ｎ Set it.

[0104] Thus, priority P is used for torque ripple compensation of each order n. ｎ When setting the parameters, if a rotational speed ω with reduced natural vibration components is used, the priority of compensation P for each order n will be ｎ The priority of each order n compensation is determined stably. That is, for example, even at or near the rotational speed ω where the priority of each order n compensation switches, oscillatory changes in priority (so-called hunting) are prevented. Therefore, the basic compensated torque T Σ Using this method, the corrected torque command value (T ２ ＊ ) is the maximum output torque T ｍａｘ Even in situations where the torque ripple exceeds a certain value, the torque ripple is reduced stably and effectively, even if the rotational speed ω of the electric motor 14 changes.

[0105] The control device for the electric motor according to the above embodiment controls the torque command value (T １ ＊ This is a control device (controller 12) for the electric motor 14 that compensates for torque ripple by correcting the fundamental amplitude K of the torque ripple for each of the multiple orders n. ｎ The basic amplitude calculation unit 31 calculates the basic amplitude K for each order n, and the basic amplitude K ｎ By correcting each of these according to the rotational speed ω of the electric motor 14, the correction amplitude K of each order n is obtained. ｎ An amplitude correction unit 35 calculates ', and the corrected amplitude K of each order n. ｎ Compensation torque T using 'Σ A compensation torque calculation unit 37 that calculates ′, and this compensation torque T Σ ′ is used to correct the torque command value (T １ ＊ ), and a torque command value correction unit 22.

[0106] Thus, the controller 12 uses the compensation torque T １ ＊ ′ used for correcting the torque command value (T Σ ′, which is corrected for each order n according to the rotational speed ω of the electric motor 14, to calculate. Therefore, even in a situation where the corrected torque command value (T ｎ ′) exceeds the maximum output torque T Σ when using the basic compensation torque T ２ ＊ ), the controller 12 can satisfactorily reduce torque ripple. In particular, by multiplying the basic compensation torque T ｍａｘ by a gain g and uniformly limiting the compensation of torque ripple for each order n, the corrected torque command value (T Σ can be made to fall within the maximum output torque T ２ ＊ ), and the controller 12 can still satisfactorily reduce torque ripple compared to the case (comparative example) where the corrected torque command value (T ｍａｘ ) is kept below the maximum output torque T.

[0107] As described above, the embodiments and modified examples of the present invention have been explained. However, the configurations described in the above embodiments and modified examples merely show a part of the application examples of the present invention and are not intended to limit the technical scope of the present invention.

Claims

1. A method for controlling an electric motor that compensates for torque ripple by correcting the torque command value, comprising: calculating the fundamental amplitude of the torque ripple for each of a plurality of orders; correcting the fundamental amplitude for each order according to the rotational speed of the electric motor to calculate the corrected amplitude for each order; calculating a compensating torque using the corrected amplitude for each order; and correcting the torque command value using the compensating torque.

2. A method for controlling an electric motor according to claim 1, comprising: calculating a correctable amount which is the deviation between the maximum output torque of the electric motor and the torque command value; setting a priority for compensating the torque ripple of each order according to the rotational speed; and determining the magnitude of the compensation amplitude of each order such that when the sum of the fundamental amplitudes of each order exceeds the correctable amount, the torque ripple of the order with the higher priority within the range of the correctable amount is preferentially compensated.

3. A method for controlling an electric motor according to claim 2, wherein the correctable amounts are used in order of increasing priority, and the corrected amplitudes of orders exceeding the correctable amounts are made smaller than the basic amplitude according to the priority.

4. A method for controlling an electric motor according to claim 2, wherein the priority of each order is set to be maximum at the natural angular frequency closest to the angular frequency of each order, and the rotational speed decreases as it moves away from the natural angular frequency.

5. A method for controlling an electric motor according to claim 2, wherein the priority is set such that the lower the rotational speed, the higher the order of the torque ripple is given priority in compensation.

6. A method for controlling an electric motor according to claim 2, comprising: reducing the natural vibration component from the rotational speed using a filter; and setting the priority using the rotational speed from which the natural vibration component has been reduced.

7. A control device for an electric motor that compensates for torque ripple by correcting the torque command value, comprising: a basic amplitude calculation unit that calculates the basic amplitude of the torque ripple for each of a plurality of orders; an amplitude correction unit that calculates a corrected amplitude for each order by correcting the basic amplitude for each order according to the rotational speed of the electric motor; a compensating torque calculation unit that calculates a compensating torque using the corrected amplitude for each order; and a torque command value correction unit that corrects the torque command value using the compensating torque.

Images