Motor control device

Abstract

According to the present invention, a processor determines a harmonic control command value (φ*dqh) from a torque command value (τ*) (harmonic control command output unit 8). The processor determines a response characteristic compensation value (Δφ*dqh) from the harmonic control command value and an electrical angular velocity (ωe) (response characteristic compensation unit 92). The processor determines a feedforward voltage command value from the harmonic control command value and the electrical angular velocity (feedforward voltage calculation unit 91). The processor determines a voltage command value (v*dq) on the basis of a control command value (φ*dq) corresponding to a torque command value 10 and the harmonic control command value (voltage command calculation unit 3). The processor switches between a first control, in which the feedforward voltage command value (v*dqh) is superimposed on the voltage command value, and a second control, in which the harmonic control command value is compensated by the response characteristic compensation value (switching unit 93), according to the switching condition.

Description

Motor control device

[0001] The present invention relates to a motor control device.

[0002] Motors are becoming increasingly powerful for use in applications with limited loading space, such as automobiles. This has led to motors being used in an area where magnetic saturation occurs. In the magnetic saturation area, torque ripple increases, inducing vibration and noise, so torque ripple suppression control is particularly necessary.

[0003] Torque ripple suppression control involves passing current harmonics that generate torque in the opposite phase to the torque ripple that occurs. To achieve this, the required inverter output voltage is calculated using feedforward (FF) from the harmonic current command value. At this time, an inverse model of the motor is required to calculate the required voltage command from the harmonic current command value, and the accuracy of achieving harmonic current is determined by the accuracy of the motor inverse model. For example, the inverse model used in Patent Document 1 considers the effects of the motor's inductance, resistance, and induced voltage, and calculates the voltage command value from the harmonic current command value.

[0004] When using an inverse model to achieve harmonic currents, it takes time to adjust the parameters of the inverse model, so some of the parameters used in the motor inverse model may be omitted. Even in this case, there is almost no error as long as the harmonic frequency is above a certain level. Therefore, the required current harmonics are achieved by FF in the high-speed range, and by current FB (feedback) in the low-speed range, where the harmonic current command value is superimposed on the current command value.

[0005] Patent No. 5574790

[0006] If the FB control band is lower than the threshold frequency where the error in the feedforward voltage calculated from the parameter-omitted inverse model of the motor becomes sufficiently small, a region will emerge where neither FF nor FB can achieve accuracy. To prevent this, the FB control band can be transparently expanded up to the threshold frequency by superimposing a harmonic command value whose phase and amplitude correspond to the response characteristics of the FB control. However, if you try to suppress torque ripple by superimposing these FF and FB simultaneously, the two will interfere with each other, making it impossible to suppress torque ripple.

[0007] An object of the present invention is to provide a motor control device that can maintain torque ripple suppression accuracy over a wide range of rotation speeds while avoiding interference between feedforward and feedback.

[0008] In order to achieve the above object, the present invention provides a motor control device that drives a motor using an inverter, and includes a processor that determines a harmonic control command value from a torque command value, determines a response characteristic compensation value from the harmonic control command value and an electrical angular velocity, determines a feedforward voltage command value from the harmonic control command value and the electrical angular velocity, determines a voltage command value based on a control command value corresponding to the torque command value and the harmonic control command value, and switches between a first control that superimposes the feedforward voltage command value on the voltage command value and a second control that compensates for the harmonic control command value with the response characteristic compensation value according to a switching condition.

[0009] According to the present invention, it is possible to maintain the accuracy of torque ripple suppression over a wide range of rotation speeds while avoiding interference between feedforward and feedback. Problems, configurations, and effects other than those described above will become apparent from the following description of the embodiments.

[0010] Fig. 1 is a schematic block diagram of a motor control device according to a first embodiment of the present invention; Fig. 2 is a schematic block diagram of a switching unit according to the first embodiment of the present invention; Fig. 3 is a schematic block diagram of a motor control device according to a second embodiment of the present invention; Fig. 4 is a schematic block diagram of a switching unit according to the second embodiment of the present invention.

[0011] 1 is a functional block diagram showing the configuration of a motor control device according to a first embodiment of the present invention. The motor control device suppresses torque ripples generated from a motor during motor control.

[0012] In the following description, the term "dq axes" refers to the "d axis and q axis." Furthermore, with respect to the subscripted parameters X (such as "i" and "v") described below, unless otherwise specified, "Xdq" refers to a "vector quantity (Xd, Xq)" and "Xuvw" refers to a "vector quantity (Xu, Xv, Xw)." Here, "uvw" refers to the three phases of AC, i.e., the "U phase, V phase, and W phase."

[0013] The motor control device 1 controls the speed and torque of the motor 5 by controlling the switching of the inverter 4 that supplies AC power to the motor 5 .

[0014] In this embodiment, the motor control device 1 is configured with an arithmetic processing device such as a microcomputer, a gate driver, etc. The arithmetic processing device includes a processor, a memory, an input / output circuit, etc., and functions as each unit by executing a predetermined program.

[0015] In this embodiment, the inverter 4 is a three-phase inverter having a three-phase full-bridge circuit as its main circuit, which is made up of semiconductor switching elements such as IGBTs and MOSFETs.

[0016] In this embodiment, the motor 5 is a three-phase AC synchronous motor, which is a rotating machine, such as a permanent magnet synchronous motor.

[0017] The motor 5 is not limited to a synchronous machine, but may be an induction machine. The motor 5 is also not limited to a rotating machine, but may be a linear motor. The motor 5 may have a power generation function.

[0018] As shown in FIG. 1 , the motor control device 1 includes a torque command value 10 (τ*), a control command output unit 2, a voltage command calculation unit 3, an electrical angle detection unit 6, an electrical angular velocity calculation unit 7, a harmonic control command output unit 8, a determination unit 9, a voltage command value dq / three-phase axis converter 11, a measurement current three-phase / dq axis converter 12, and a current-flux converter 13.

[0019] The torque command value 10 is given from a higher-level controller or operation system.

[0020] The control command output unit 2 calculates the dq-axis magnetic flux command value (φ* ｄｑ ) is output. Generally, it is determined as a magnetic flux value corresponding to a current value that maximizes the torque output per current (MTPA), but the present invention is effective for any magnetic flux command value that realizes a torque command. The magnetic flux command value may be calculated from the torque command, or may be determined by referring to a preset table.

[0021] The voltage command calculation unit 3 calculates the d-axis and q-axis magnetic flux detection values ​​φ ｄｑ and the dq-axis magnetic flux command value φ** after harmonics are superimposed ｄｑ Depending on the difference between ｄｑ is φ* ｄｑ The voltage command value v* is set to ｄｑ Calculate.

[0022] The electrical angle detector 6 detects the electrical angle θ based on a rotation position signal from a rotation sensor (not shown) provided in the motor 5. ｅ The rotation sensor calculates and outputs the mechanical angle (θ ｍ ), the electrical angle detector 6 detects the number of pole pairs (p) of the motor 5 and the ｅ and θ ｍ and the relationship between p (θ ｅ = p θ ｍ ) and θ ｅ Calculate.

[0023] The rotation sensor may be a mechanical angle resolver, a hall sensor, a rotary encoder, etc. Alternatively, a general position sensorless control method may be used to estimate the electrical angle from the current or voltage.

[0024] The electrical angle velocity calculation unit 7 calculates the electrical angle velocity ω based on a rotation position signal from a rotation sensor (not shown) provided in the motor 5, similarly to the electrical angle detection unit 6 (electrical angle calculation unit). ｅ The electrical angular velocity calculation unit 7 calculates and outputs the number of pole pairs (p) of the motor 5 and the electrical angle θ ｅ and mechanical angle θ ｍ and the relationship between p (θ ｅ = p θ ｍ ) and θ ｅ and ω ｅ The relationship between (ω ｅ = dθ ｅ / dt) to obtain ω ｅ The electrical angular velocity calculation unit 7 calculates the θ ｅ Differentiating ω ｅ It may also be comprised of a differentiator that calculates:

[0025] The harmonic control command output unit 8 outputs a harmonic control command value φ* in accordance with τ*. dqhOutput is φ* dqh may be calculated from the torque command or may be found by referring to a preset table.

[0026] The determination unit 9 includes a feedforward voltage calculation unit 91 , a response characteristic compensation unit 92 , and a switching unit 93 .

[0027] The feedforward voltage calculation unit 91 calculates φ* dqh and electrical angular velocity ω ｅ The FF voltage v*dqh is calculated and output based on equation (1). Note that s in equation (1) is the Laplace operator.

[0028]

[0029] Here, we will explain how to derive equation (1). The inverse model of a general motor on the dq axes is given by equation (2).

[0030]

[0031] However, in equation (2), Ldq and Lqd represent the interference inductance between the d and q axes, and Kedh and Keqh represent the harmonic components of the d and q axes of the magnetic flux of the magnet. Variables with a t at the end of the L subscript, which represents inductance, represent dynamic inductance, while variables without a t represent static inductance. Static inductance is a value determined by the ratio of the d and q axis currents to the d and q axis magnetic flux, and dynamic inductance refers to the rate of change in magnetic flux near the reference current. Here, we use the relationship between the magnetic flux defined in equation (3) and the dynamic inductance shown in equation (4).

[0032]

[0033]

[0034] By transforming equation (2) using equations (3) and (4) and extracting only the harmonic components, equation (5) is obtained.

[0035]

[0036] Here, the second and third terms on the right side of equation (5) increase in proportion to the frequency of the harmonic control command value, so above a certain threshold frequency the influence of the first term on the right side becomes small enough to be ignored, yielding equation (1). Conversely, in the low-speed region, the influence of the first term on the right side is large, making it impossible to calculate the correct FF voltage, so FB control must be effective up to the threshold frequency.

[0037] The response characteristic compensation unit 92 includes a response characteristic compensation table or a response characteristic compensation function, and calculates the electrical angular velocity ω ｅ and the harmonic control command value φ* by the amount of correction of the phase θmod and amplitude Imod calculated from the table or function based on the order of the harmonic. dqh The harmonic control command value difference Δφ* when dqh where n is the order of the harmonic, such as 6 or 12, which is the order of the target for torque ripple suppression, and φ* dqnh indicates the n-th harmonic control command value component.

[0038]

[0039] As mentioned above, since the threshold frequency is determined by the motor constant, if the feedback band is equal to or lower than this frequency, the torque ripple suppression control will not operate sufficiently in the region from the upper limit frequency of the feedback band to the threshold frequency. Here, the response characteristic compensator 92 corresponds to an inverse function of the transfer characteristic of the feedback control, and by applying response characteristic compensation to the harmonic control command value, it becomes possible to expand the feedback control band only for harmonics.

[0040] As shown in FIG. 2, the switching unit 93 is composed of a voltage command value calculation 931, a harmonic component calculation 932, and a difference comparison 933. dq and φ* dq and φ* dqh and Δφ* dqh Based on v*dqh, Δφ* dqh At least one of the outputs v*dqh and v*dqh is set to 0.

[0041] The operation and function of the switching unit 93 will be described below.

[0042] First, the operation when the speed increases from low to high will be described. At low speed, v*dqh is set to 0 by the difference comparison 933. The voltage command value calculation 931 calculates Δφ* dqh The same calculation as that of the voltage command calculation unit 3 is performed on the result of adding the above. A harmonic component calculation unit 932 extracts and outputs the harmonic component of the nth electrical angle from the output components of 931. A difference comparison unit 933 compares Δφ* with the output value of 932 when the frequency of the harmonic increases and the output value of 932 and the output of the feedforward calculation unit match within a certain error. dqh The output of the feedforward calculation unit is set to 0, and the value of the feedforward calculation unit is output as v*dqh.

[0043] Next, the operation when the speed decreases from high to low will be described. At high speed, Δφ* dqh is set to 0 by the difference comparison 933. The harmonic component calculation 932 extracts and outputs the harmonic component of the nth electrical angle from the harmonic component 930. When the frequency of the harmonic falls and the output value of the difference comparison 933 exceeds a certain value, Δφ* dqh and outputs 0 as v*dqh.

[0044] When the switching unit 93 is not present, Δφ* dqh and v*dqh are superimposed at the same time. At this time, Δφ* dqh The transfer characteristics of the FB control used to calculate v*dqh change from the design value when v*dqh is superimposed, causing the two controls to interfere with each other and preventing the torque ripple suppression control from operating properly.

[0045] The voltage command value dq / three-phase axis converter 11 converts the voltage command value v* output from the voltage command calculation unit 3 into ｄｑ θ ｅ The inverter 4 generates a signal for controlling the switching of the inverter 4 by converting the voltage into a three-phase voltage according to θ ｅ Instead of θ, a delay is compensated for until the voltage command value is actually reflected in the inverter. ｅc may also be used.

[0046] The measurement current three-phase / dq axis converter 12 converts the three-phase current detected by the current sensor into θ ｅ The dq axis current i ｄｑ In this case, θ ｅInstead of θ, which takes into account the current detection delay ｅd may also be used.

[0047] The main features of the first embodiment can be summarized as follows.

[0048] The motor control device 1 drives a motor 5 by an inverter 4. The processor of the motor control device 1 calculates a harmonic control command value (φ*) from a torque command value 10 (τ*). dqh ) is determined (harmonic control command output unit 8). dqh ) and electrical angular velocity (ω ｅ ) to the response characteristic compensation value (Δφ* dqh ) (response characteristic compensation unit 92). The processor determines the harmonic control command value (φ* dqh ) and electrical angular velocity (ω ｅ The processor determines a feedforward voltage command value (v*dqh) from the torque command value τ* (feedforward voltage calculation unit 91). ｄｑ ) and harmonic control command value (φ* dqh ) and the voltage command value (v* ｄｑ The processor determines the voltage command value (v*) according to the switching conditions (voltage command calculation unit 3). ｄｑ ) and the first control superimposes the feedforward voltage command value (v*dqh) on the response characteristic compensation value (Δφ* dqh ) is the harmonic control command value (φ* dqh ) and the second control that compensates for the above-mentioned difference (switching unit 93).

[0049] By switching exclusively between the first control and the second control, the first control and the second control do not interfere with each other. The feedforward voltage of the first control causes harmonic currents to flow, generating torque in the motor that suppresses torque ripple. Furthermore, since the feedforward voltage does not become a disturbance during response characteristic compensation of the second control, the accuracy of torque ripple suppression does not decrease. This makes it possible to maintain torque ripple suppression accuracy over a wide rotation speed range while avoiding interference between feedforward and feedback. As a result, motor noise and vibration can be suppressed. The values ​​are determined, for example, by calculating values ​​using a formula or deriving values ​​from a table.

[0050] The processor may, for example, calculate a voltage command value (v* ｄｑ The first control (FF voltage superposition) and the second control (response characteristic compensation) are switched depending on the difference between the harmonic components (output of the harmonic component calculation 932) of the output voltage (v*dqh) and the feedforward voltage command value (v*dqh).

[0051] In the first embodiment, the frequency of the harmonics increases, and the voltage command value (v* ｄｑ When the harmonic components of the output voltage (v*dqh) match within a certain error, the control is switched from the second control (response characteristic compensation) to the first control (FF voltage superposition).

[0052] As a result, for example, it is possible to execute the more accurate control of the first control (FF voltage superposition) and the second control (response characteristic compensation) with the frequency at which the harmonic component of the voltage command value and the feedforward voltage command value become equal as the boundary. Note that in the first embodiment, the first control (FF voltage superposition) is executed in a region where the harmonic frequency is high, and the second control (response characteristic compensation) is executed in a region where the harmonic frequency is low.

[0053] The processor calculates the harmonic control command value (φ* dqh ) and the same physical quantity as the measurement and control quantity (φ ｄｑ ) and harmonic control command value (φ* dqh ) the first control (FF voltage superposition) and the second control (response characteristic compensation) may be switched over depending on the difference between the first control (FF voltage superposition) and the second control (response characteristic compensation).

[0054] This makes it possible to execute the more accurate control of the first control (FF voltage superposition) and the second control (response characteristic compensation), for example, with the frequency at which the difference between the measured control amount and the harmonic control command value becomes a predetermined value as the boundary.

[0055] The processor calculates the harmonic control command value (φ* dqh ) contains multiple frequency components, the harmonic control command value (φ* dqh The first control (FF voltage superposition) and the second control (response characteristic compensation) are switched for each frequency component (nth-order harmonic component of electrical angle) included in the frequency component (nth-order harmonic component of electrical angle). This makes it possible to maintain the accuracy of torque ripple suppression for each frequency component.

[0056] The processor calculates a response characteristic compensation value (Δφ* dqh ) is the harmonic control command value (φ* dqh ) is subjected to gain compensation and phase compensation. This results in, for example, a good transient response to harmonic currents.

[0057] In the first embodiment, the control command value (φ* ｄｑ ) and harmonic control command value (φ* dqh The physical quantity of the magnetic flux is magnetic flux, but it may also be current. When the physical quantity is magnetic flux, the number of motor constants used in the calculation of the feedforward voltage command value can be reduced compared to when the physical quantity is current.

[0058] Control command value (φ* ｄｑ ) is, for example, a magnetic flux command value that causes the motor 5 to output a torque corresponding to the torque command value 10(τ*). dqh ) is, for example, a magnetic flux command value that causes the motor 5 to generate a torque that is in the opposite phase to the torque ripple. dqh ) superimposed on the control command value (φ* ｄｑ ) and the measured magnetic flux value (φ ｄｑ ) and the response characteristic compensation value (Δφ* dqh ) is added to the voltage command value (v* ｄｑ The first control (FF voltage superposition) and the second control (response characteristic compensation) are switched depending on the difference between the harmonic components of the output voltage (v*dqh) and the feedforward voltage command value (v*dqh).

[0059] This makes it possible to execute the more accurate control of the first control (FF voltage superposition) and the second control (response characteristic compensation), for example, using the frequency at which the accuracy of the harmonic components of the voltage command value due to response characteristic compensation and the accuracy of the feedforward voltage command value become equal as the boundary.

[0060] The processor calculates the control command value (φ* ｄｑ ) and the measurement control amount (φ ｄｑ ) so that the voltage command value (v* ｄｑ ) is determined. This allows feedback control of harmonic currents.

[0061] Second Embodiment The difference from the first embodiment will be explained using the block diagram of the second embodiment shown in Fig. 3. In this embodiment, the configuration of the determination unit 9 is different, and a switching unit 93 as shown in Fig. 4 is provided.

[0062] The switching unit 93 is composed of a threshold frequency setting unit 934 , a comparator 935 , and a harmonic control command switching unit 936 .

[0063] The threshold frequency setting 934 is set as a value that allows for an allowable error between equations (1) and (5). This value may be a fixed value, or it may be a function or table that has variables that affect the first term of equation (5), such as the torque command value or the current value, thereby enabling the frequency at which switching should be performed to be specified more accurately.

[0064] A comparator 935 compares the electrical angle frequency for each order with the threshold frequency, and when the threshold frequency is higher, a harmonic control command switch 936 outputs φ* to the response characteristic compensation unit for each order. dqnh and outputs 0 to the feedforward voltage calculation unit, otherwise, outputs 0 to the response characteristic compensation unit and outputs φ* to the feedforward voltage calculation unit. dqnh Determines the state to output.

[0065] Furthermore, by providing hysteresis characteristics to the harmonic control command switch 936, the influence of chattering can be suppressed.

[0066] The main features of the second embodiment can be summarized as follows.

[0067] The processor calculates the electrical angular velocity (ω ｅIn the second embodiment, the processor switches between the first control (FF voltage superposition) and the second control (response characteristic compensation) based on the electrical angular velocity (ω ｅ If the electrical angular velocity (ω ｅ ) is smaller than the threshold value, the second control (response characteristic compensation) is performed. As a result, for example, the first control (FF voltage superposition) is performed in a region where the electrical angular velocity is high, and the second control (response characteristic compensation) is performed in a region where the electrical angular velocity is low.

[0068] In the second embodiment, the switching condition has a hysteresis characteristic, which can suppress chattering.

[0069] The present invention is not limited to the above-described embodiments, but includes various modifications. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those including all of the described configurations. Furthermore, it is possible to replace part of the configuration of one embodiment with the configuration of another embodiment, or to add the configuration of another embodiment to the configuration of one embodiment. Furthermore, it is possible to add, delete, or replace part of the configuration of each embodiment with other configurations.

[0070] Furthermore, some or all of the above-described configurations, functions, etc. may be implemented in hardware, for example, by designing them as integrated circuits. Furthermore, the above-described configurations, functions, etc. may be implemented in software by a processor interpreting and executing a program that implements each function. Information such as the programs, tables, and files that implement each function can be stored in a memory, a recording device such as a hard disk or SSD (Solid State Drive), or a recording medium such as an IC card, SD card, or DVD.

[0071] The embodiment of the present invention may be configured as follows.

[0072] (A1) A motor control device that drives a motor using an inverter, comprising: a harmonic control command calculation unit that calculates a harmonic control command value from a torque command; a response characteristic compensation unit that calculates a response characteristic compensation value from the harmonic control command value and its frequency; a feedforward voltage calculation unit that calculates a feedforward voltage command value from the harmonic control command value; a voltage command calculation unit that calculates a voltage command value based on a control command corresponding to the torque command; and a switching unit that switches between superimposing the feedforward voltage command value on the voltage command value and calculating the voltage command value based on an output value of the response characteristic compensation unit, depending on the harmonic control command value.

[0073] (A2) The motor control device according to (A1), wherein the switching unit performs switching based on a difference between harmonic components of the voltage command value and the feedforward voltage command value.

[0074] (A3) The motor control device according to (A1), wherein the switching unit switches based on the difference between the harmonic control command value and a measured control amount having the same physical amount as the harmonic control command value.

[0075] (A4) The motor control device according to (A1), wherein the switching unit switches depending on the frequency of the harmonic control command value.

[0076] (A5) The motor control device according to any one of (A2) to (A4), wherein the switching unit has a hysteresis characteristic with respect to the switching condition.

[0077] (A6) The motor control device according to any one of (A2) to (A5), wherein the switching unit switches for each frequency when the harmonic control command value includes a plurality of frequency components.

[0078] According to (A1)-(A6), the torque ripple suppression effect can be ensured over a wide rotation speed range without significantly extending the adaptation time.

[0079] DESCRIPTION OF SYMBOLS 1...Motor control device 2...Control command output section 3...Voltage command calculation section 4...Inverter 5...Motor 6...Electrical angle detection section 7...Electrical angular velocity calculation section 8...Harmonic control command output section 9...Determination section 91...Feedforward voltage calculation section 92...Response characteristic compensation section 93...Switching section 10...Torque command value

Claims

1. A motor control device that drives a motor using an inverter, comprising a processor that determines a harmonic control command value from a torque command value, determines a response characteristic compensation value from the harmonic control command value and an electrical angular velocity, determines a feedforward voltage command value from the harmonic control command value and the electrical angular velocity, determines a voltage command value based on a control command value corresponding to the torque command value and the harmonic control command value, and switches between a first control that superimposes the feedforward voltage command value on the voltage command value and a second control that compensates for the harmonic control command value with the response characteristic compensation value depending on a switching condition.

2. A motor control device according to claim 1, characterized in that the processor switches between the first control and the second control depending on the difference between the harmonic components of the voltage command value and the feedforward voltage command value.

3. A motor control device as described in claim 1, characterized in that the processor switches between the first control and the second control depending on the difference between a measured control amount of the same physical quantity as the harmonic control command value and the harmonic control command value.

4. A motor control device according to claim 1, characterized in that the processor switches between the first control and the second control depending on the electrical angular velocity.

5. A motor control device according to claim 1, characterized in that the switching condition has a hysteresis characteristic.

6. A motor control device as described in claim 1, characterized in that, when the harmonic control command value includes a plurality of frequency components, the processor switches between the first control and the second control for each frequency component included in the harmonic control command value.

7. A motor control device according to claim 1, characterized in that the processor performs gain compensation and phase compensation on the harmonic control command value using the response characteristic compensation value in the second control.

8. A motor control device according to claim 1, wherein the physical quantities of the control command value and the harmonic control command value are magnetic flux or current.

9. A motor control device as described in claim 1, wherein the control command value is a magnetic flux command value that causes the motor to output a torque corresponding to the torque command value, and the harmonic control command value is a magnetic flux command value that causes the motor to generate a torque in the opposite phase to the torque ripple, and the processor switches between the first control and the second control based on the difference between the harmonic component of the voltage command value, which is determined by adding the response characteristic compensation value to the difference between the control command value superimposed with the harmonic control command value and the measured magnetic flux value, and the difference between the harmonic component of the voltage command value and the feedforward voltage command value.

10. A motor control device according to claim 1, wherein said processor determines said voltage command value so that said control command value coincides with a measured controlled variable.

Images