Motor torque ripple suppression method based on integral proportion and quasi-proportional resonance control

Abstract

The invention discloses a motor torque ripple suppression method based on integral proportion and quasi-proportional resonance control, and belongs to the technical field of motor control, and the method comprises the following steps: S1, obtaining the actual position of a motor rotor through the sampling of a sensor, and calculating the rotating speed n of the motor rotor; s2, the difference between the current rotating speed and the target rotating speed is obtained, and idref and iqref needed by current control are obtained through an integral proportion and quasi-proportional resonance comprehensive controller; and S3, performing subtraction between the idref and the id and between the iqref and the iq, obtaining uq and ud through a PI controller, inputting the uq and ud into a space vector modulation module after inverse Park conversion, and outputting a PWM control signal to control the motor. According to the invention, the torque pulsation with known frequency can be effectively inhibited, the speed pulsation is reduced by about 82%, and the stability of a speed ring is obviously improved; the response speed is obviously improved, and the aim of optimization design is achieved.

Description

Technical Field

[0001] The invention belongs to the technical field of motor control and relates to a method for suppressing torque pulsation of a servo motor based on integral proportion and quasi-proportional resonance control. Background Art

[0002] A servo motor is a motor whose rotor excited by permanent magnets rotates synchronously with the stator's spatial magnetic field. Due to its advantages such as high efficiency and high power density, servo motors are widely used in key areas such as advanced rail transportation equipment, robots, and aerospace equipment. However, the servo motor system itself has a large amount of interference such as nonlinear friction, cogging torque, and permanent magnet flux harmonics, which cause the torque on the motor shaft to pulsate, thereby reducing the running stability and control accuracy of the motor. Traditional speed control methods are difficult to achieve excellent control effects. Therefore, suppressing torque pulsation has always been a hot topic of research at home and abroad.

[0003] Since the torque pulsation in the actual system is difficult to model or measure directly, an observer can be designed to observe the torque pulsation and feed the observed value forward to the control system for compensation. For example, the literature (Ji Wengang, Lin Li, Dai Fengyan. Nonlinear Compensation Algorithm for Eliminating Low Speed ​​Stick-Slipin Servo System [C] / / 2009 Second International Conference on Intelligent Computation Technology and Automation, 2009: 868-871.) proposed a strategy for dynamically adjusting the PI controller gain according to the working point based on the Stribeck static friction model, which effectively suppressed the influence of friction torque. This type of observer-based anti-disturbance control strategy is also called active anti-disturbance control. In contrast, the strategy of suppressing torque pulsation by only designing a feedback controller with strong anti-disturbance capability without using an observer to observe the pulsation information is called passive anti-disturbance control. For example, the literature (Zhang Xiaoyu. Discussion on "Torque Ripple Minimization of PMSM Based on Robust ILC via Adaptive Sliding Mode Control" [J]. IEEE Access, 2019, 7: 94899-94906.) adopts an iterative learning control algorithm to suppress the influence of periodic torque pulsation without obtaining an accurate mathematical model of the system.

[0004] CN112039386A discloses a method for suppressing torque pulsation of a permanent magnet synchronous motor based on fuzzy quasi-proportional resonance, which uses a fuzzy quasi-proportional resonance controller to replace a traditional PI controller, sets a current controller for suppressing harmonics, and realizes rapid control of a permanent magnet synchronous motor with high-order, nonlinear characteristics, thereby improving the motor response performance.

[0005] CN114123874A discloses a commutation torque pulsation suppression method, storage medium and device based on off-phase current control, wherein the motor speed ω is calculated according to the commutation signal, and the back electromotive force E is obtained according to the motor speed ω; then the lag time t is obtained based on the back electromotive force E d , and according to the motor speed ω and the lag time t d Get the commutation signal lag angle θ d ; When the commutation signal is detected, the commutation signal is delayed by θ d , to achieve the suppression of commutation torque pulsation.

[0006] CN114584024A discloses a method for suppressing torque pulsation during fault-tolerant operation of a dual three-phase permanent magnet synchronous motor. The method disconnects the corresponding harmonic sub-plane current loop according to the corresponding phase where an open circuit fault occurs, and replaces it with a set corresponding feedforward signal. A Quasi-PR controller is proposed to complete the switching of the fault-tolerant algorithm and realize the suppression of torque pulsation during fault-tolerant operation.

[0007] CN114665771A discloses a method for suppressing torque pulsation of an electrically excited double-pole motor. The method adopts a torque inner loop to replace a current inner loop, and adds an excitation current regulation link. The excitation current and the armature current are coordinated to directly control the motor output torque, thereby suppressing the torque pulsation problem caused by insufficient motor output during commutation using the existing direct instantaneous torque control method.

[0008] CN116566261A discloses a method for suppressing torque pulsation of a stator-joined dual-axis permanent magnet motor, the method comprising: extracting the torque pulsation of the joined motor according to the rotation speed of the joined motor and calculating the q-axis harmonic injection current required by the harmonic current injection algorithm; according to the torque pulsation characteristics of the joined motor caused by the magnetic force of the magnetic poles, it is proposed to use a random approximation algorithm to optimize the phase and amplitude of the harmonic injection current in real time; superimposing the optimized harmonic injection current in front of the current loop, so that the motor generates a corresponding electromagnetic torque to offset the torque pulsation generated by the magnetic force of the magnetic poles, so that the joined motor can run smoothly.

[0009] When a servo system based on a servo motor runs at low speed, due to external disturbances, the enhancement of nonlinear effects in the system and the limitation of the position sensor resolution, the motor torque pulsates greatly, affecting the stability of the system at low speed. It is necessary to develop targeted torque pulsation suppression methods to improve the suppression effect of torque pulsation, optimize the tracking performance of the servo motor during operation, especially at low speed, reduce speed pulsation, and improve the steady-state and dynamic performance of the servo system at low speed. Summary of the invention

[0010] The purpose of the present invention is to provide a motor torque pulsation suppression method based on integral proportion (IP) and quasi-proportional resonance (QPR) control, and to propose a comprehensive control strategy combining an integral proportion (IP) controller with a quasi-proportional resonance (QPR) controller for periodic torque pulsation caused by load fluctuations, friction torque changes, motor cogging torque, electromagnetic wave torque and inverter dead zone effect at low speed, which significantly improves the torque pulsation suppression effect compared with traditional control schemes. The purpose of the present invention is achieved through the following specific technical solutions.

[0011] The method for suppressing motor torque pulsation based on integral proportional and quasi-proportional resonant control comprises the following steps:

[0012] S1: The actual position of the motor rotor is obtained through sensor sampling to calculate its speed n;

[0013] S2: Difference between the current speed and the target speed, and obtain the dq axis stator current i required for current control through the integral proportional and quasi-proportional resonant integrated controller. dref with i qref ;

[0014] S3: i dref with i d 、i qref with i q Make a difference and get u through PI controller q with u d , and then after the inverse Park transformation, it is input into the space vector modulation module, and the PWM control signal is output to control the motor; where i d 、i q are the stator currents of the dq axis, u d 、u q are the stator voltages of the dq axes respectively.

[0015] Furthermore, in step S1, the position of the rotor is sampled using a sensor, and the rotor speed n is calculated by the position difference at each time Δt. In this way, the position and speed can be collected using only a position sensor, which reduces the cost.

[0016] Further, in step S3, the three-phase current i is sampled and obtained. a 、i b 、i c , the sampled three-phase current is subjected to Clark transformation and Park change to obtain i q with i d .

[0017] Furthermore, in step S2, a parallel quasi-proportional resonant controller is used as part of the speed loop, and its transfer function is:

[0018]

[0019] Where K p is the proportionality coefficient, ω c is the cut-off angular frequency, ω is the resonant angular frequency, and K is the resonant parameter.

[0020] The use of a parallel quasi-proportional resonant controller can suppress the harmonics of both frequencies at the same time.

[0021] Furthermore, in step S2, based on the parallel control of the integral proportional and quasi-proportional resonant controllers, the parameters of the two controllers are adjusted separately; the integral proportional controller is adjusted using a zero-pole adjustment method, and the bandwidth is selected from 1 / 10 to 1 / 5 of the current loop; the dynamic response capability of the controller can be optimized while maintaining the parallel proportional resonant controller's ability to suppress harmonics.

[0022] Compared with the prior art, the present invention has the following beneficial technical effects: In view of the periodic torque pulsation existing in the operation of the servo motor, the present invention proposes an IP+QPR integrated controller and optimizes the speed loop in the motor vector control. Simulation verification shows that compared with the traditional PI controller, the controller can effectively suppress the torque pulsation of known frequency, reduce the speed pulsation by about 82%, and significantly improve the stability of the speed loop. Compared with the use of the QPR controller alone, the IP+QPR integrated controller also significantly improves the response speed, achieving the goal of optimized design. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is the speed closed-loop control block diagram.

[0024] Figure 2 This is a schematic diagram of the ideal servo motor vector control speed closed-loop structure.

[0025] Figure 3 This is the block diagram of the QPR resonant controller.

[0026] Figure 4 It is the amplitude-frequency and phase-frequency characteristics of the PR controller.

[0027] Figure 5It is the amplitude-frequency and phase-frequency characteristics of the QPR controller.

[0028] Figure 6 It is the open-loop amplitude-frequency and phase-frequency characteristics of the PR controller speed loop.

[0029] Figure 7 It is the open-loop amplitude-frequency and phase-frequency characteristics of the speed loop of the QPR controller.

[0030] Figure 8 This is the structure diagram of the IP controller.

[0031] Fig. 9 This is the structural block diagram of the IP controller after the transformation.

[0032] Fig.10 It is the open-loop Bode plot of the parallel QPR controller.

[0033] Fig.11 It is the open-loop Bode diagram of the PI+QPR integrated controller.

[0034] Fig.12 It is a schematic diagram of the vector control speed closed loop structure using the IP+QPR integrated controller.

[0035] Fig.13 These are some speed-time curves of PI controller and QPR controller.

[0036] Fig.14 These are some speed-time curves of IP+QPR controller and QPR controller.

[0037] Fig.15 It is the speed-time curve of the beginning part of IP+QPR controller and QPR controller. DETAILED DESCRIPTION

[0038] The technical solution of the present invention is described clearly and completely below in conjunction with the accompanying drawings of the specification. Obviously, the described implementation is only a part of the implementation of the present invention, not all of the implementations. Based on the implementation of the present invention, all other implementations obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] The present invention provides a motor torque pulsation suppression method based on integral proportion (IP) and quasi-proportional resonance (QPR) control, comprising the following steps: S1: acquiring the actual position of the motor rotor through sensor sampling to calculate its speed n; S2: subtracting the current speed from the target speed, and obtaining the dq axis stator current i required for current control through an integral proportion and quasi-proportional resonance integrated controller. dref with i qref ; S3: i dref with i d、i qref with i q Make a difference and get u through PI controller q with u d , and then after the inverse Park transformation, it is input into the space vector modulation module, and the PWM control signal is output to control the motor; where i d 、i q are the stator currents of the dq axis, u d 、u q are the stator voltages of the d and q axes respectively. The details are as follows.

[0040] The control scheme of the present invention is based on traditional vector control. Traditional vector control is to decompose the stator current of the servo motor into a stator current excitation component in the same direction as the rotor flux and a stator current torque component orthogonal to the flux direction. These two components are orthogonal to each other and are controlled by controllers respectively. The control method requires a rotating coordinate transformation. The mathematical model of the servo motor under the dq axis is:

[0041]

[0042] T e =1.5p n ψ f i a (3)

[0043]

[0044] Where u d 、u q are the stator voltages of the dq axes respectively; R s , L s are the stator resistance and stator inductance respectively; i d 、i q are the stator currents of the dq axis, ω e is the electrical angular velocity; J is the moment of inertia; ω i is the motor angular velocity; T L is the load torque; B is the friction coefficient where T e is the electromagnetic torque; ψ f is the permanent magnet rotor flux.

[0045] According to equations (1)-(4), when both the speed loop and the current loop use PI controllers, without considering the friction coefficient, the simplified speed loop closed loop is as follows: Figure 1 shown.

[0046] The speed error passes through the speed loop and the current loop to obtain the current i a, the electromagnetic torque is calculated according to formula (3), the motor electrical angular velocity can be calculated according to formula (4), and finally the motor speed is converted and fed back to the speed loop.

[0047] The structural diagram of the FOC control strategy is as follows Figure 2 As shown in the figure: the three-phase stator current sampled from the motor is transformed by Clark and Park to obtain the dq axis current for the current control loop. The real-time rotational angular velocity of the motor is obtained using a position or speed sensor, and the real-time position of the rotor is obtained using a position sensor. The angle is used for Park and anti-Park transformations, and the speed is input into the speed control loop as the sampling amount. The speed error is obtained by giving a given speed, and the speed error is used as the input of the current controller after passing through the speed controller. After the output of the current controller is transformed by anti-Park, the PWM control signal of the three-phase bridge arm is obtained through the SVPWM modulation strategy, thereby completing the closed-loop vector control system of the servo motor.

[0048] Many non-ideal factors in the servo motor control system cause periodic pulsation of the output torque. The present invention mainly proposes a suppression control scheme for the periodic torque pulsation.

[0049] 1 Analysis of Torque Pulsation Sources

[0050] (1) Permanent magnet flux harmonics

[0051] There are multiple harmonics in the permanent magnet flux in the actual motor system:

[0052]

[0053] The harmonics in phase B and phase C are completely consistent with equation (5). π / 3 electrical degree difference.

[0054] Transforming it to the dq axis coordinate system, the expression can be obtained:

[0055]

[0056] It can be seen that the frequency of the permanent magnet flux harmonic is 6npω i .

[0057] (2) Cogging torque

[0058] Cogging torque is caused by the mechanical structure of the motor body and has nothing to do with the electronic system. Therefore, cogging torque exists regardless of whether current is passed through the motor stator winding. Cogging torque is closely related to the position of the motor rotor. Researchers have shown that cogging torque is periodic and will cause torque pulsation with a frequency multiple of the motor's fundamental mechanical frequency m:

[0059] Te =Kcos(mθ+α) (8)

[0060] Where c is the greatest common divisor of 2p and the number of teeth z:

[0061]

[0062] (3) Inverter dead zone effect

[0063] In the process of pulse width modulation of voltage-type PWM inverter, in order to avoid direct short circuit of two complementary power devices in the same phase, it is necessary to add a dead time of several microseconds to the PWM pulse signal. However, adding dead time will affect the duty cycle of the pulse signal, resulting in a deviation between the output voltage of the inverter and the expected modulation voltage. The inverter dead time effect will introduce a 6n harmonic error voltage into the system.

[0064] 2PR control and QPR control

[0065] The proportional resonant controller can provide infinite gain at a specified frequency, so it has strong selectivity and responsiveness to the signal of this frequency. Utilizing this characteristic, the specific periodic torque pulsation in the operation of the motor can be effectively suppressed. However, the proportional resonant controller has the problem of low stability margin. In order to solve this problem, the present invention adopts a quasi-proportional resonant (QPR) controller. The transfer function of the controller is:

[0066]

[0067] Its structural diagram is as follows Figure 3 shown.

[0068] The transfer function of the PR controller is:

[0069]

[0070] according to Figure 1 The open-loop transfer function of the speed loop can be obtained as:

[0071]

[0072] According to the transfer functions of the QPR controller and the PR controller, the Bode diagrams of the two controllers can be drawn. Then, the transfer functions of the two controllers are brought into the speed loop open-loop transfer function, and the Bode diagrams are drawn to obtain Figures 4 to 7 .

[0073] from Figures 4 to 7It can be observed that the gain of the PR controller at the resonant frequency is much greater than that of the QPR controller, but its phase angle at the resonant frequency in the speed loop is close to -180 degrees, and the phase stability margin is extremely low. In contrast, although the gain of the QPR controller at the resonant frequency is smaller, its phase margin is greater than that of the PR controller.

[0074] 3IP Control

[0075] IP control has the same robustness as proportional integral control, and has the advantage of small overshoot when used in the speed loop, so the IP controller is used as part of the comprehensive control in the present invention. The control law of the IP controller is:

[0076]

[0077] Figure 8 is the structural block diagram of the IP controller, and G(s) is the transfer function of the controlled system.

[0078] By performing an equivalent transformation on its structural block diagram, we can obtain a block diagram such as Fig. 9 shown.

[0079] The control effect of the speed closed-loop system based on IP control is the same as that of the speed closed-loop system based on PI control by adding a low-pass filter at a given point, so it plays a control effect of reducing the overshoot.

[0080] 4IP+QPR integrated control

[0081] According to the previous analysis of the main torque pulsation of the servo motor at ultra-low speed, it can be seen that there are torque pulsations of various frequencies when the servo motor is running. In order to suppress pulsations of different periods at the same time, two QPR controllers with different resonant frequencies are connected in parallel to achieve the effect of suppressing two frequencies at the same time. The transfer function is:

[0082]

[0083] The Bode plot of the parallel QPR controller is shown in Fig.10 shown.

[0084] It can be seen from the Bode plot that when QPR controllers with different resonant frequencies are connected in parallel, effective suppression of two frequency signals can be achieved. However, it can be observed from the Bode plot that the controller has a poor suppression effect on low-frequency harmonics. In order to improve its performance in the low-frequency band, a PI controller is used in parallel with it.

[0085] The open-loop Bode diagram of the integrated controller after parallel connection is as follows: Fig.11 shown.

[0086] According to the previous analysis, the PI+QPR controller shows good suppression performance in the low frequency band. Now replace the PI controller with the IP controller to form an IP+QPR integrated controller. This controller can suppress signals of two or more frequencies at the same time, while maintaining the good dynamic performance of the PI controller, and has strong robustness, which can effectively suppress the overshoot in speed control, thereby achieving steady-state operation of the motor at low speed.

[0087] The structural diagram of the IP+QPR integrated controller is as follows Fig.12 shown.

[0088] In order to verify the torque suppression performance based on the IP+QPR controller proposed in this invention, a servo motor simulation model was built. The current loop uses a PI controller, the speed loop uses a PI controller, and the parallel QPR controller and the IP+QPR integrated controller used by the main body. The resonant frequency of the parallel QPR controller is set to the frequency of the input torque pulsation. The motor simulation parameters are: the number of pole pairs is 10 pairs, the cogging torque is considered, and the input torque pulsation is T d =0.5cos(60θ)+0.5cos(120θ), the rated speed is set to 5 rpm, which is convenient for observing the dynamic response capability and anti-interference capability of the control at the same time. The simulation results can be obtained as follows Figure 13 to Figure 15 shown.

[0089] pass Fig.13 It can be observed that there is a performance gap between PI control and QPR controller in suppressing periodic torque pulsation. When the control is stable, when the PI controller is used as the speed loop, the speed pulsates by about ±0.01 rpm, while when the QPR control is used as the speed loop, the pulsation is only about ±0.0018 rpm, which is only 18% of the PI controller. This shows that QPR control has a better effect in suppressing signals of specific frequencies. Fig.14 It can be observed that the IP+QPR controller has almost the same pulsation suppression performance as the QPR controller, with the speed pulsation of both controllers being approximately ±0.001 rpm. Fig.15 It can be observed that the QPR controller has a large overshoot at startup, and the time required to reach steady state is much longer than the time required for the IP+QPR controller to reach steady state. The simulation results show that the IP+QPR controller has excellent response performance, which verifies the effectiveness of the controller.

[0090] Aiming at the periodic torque pulsation of servo motors during operation, the present invention proposes an IP+QPR integrated controller and optimizes the speed loop in the motor vector control. Simulation verification shows that compared with the traditional PI controller, the controller can effectively suppress the torque pulsation of known frequency, reduce the speed pulsation by about 82%, and significantly improve the stability of the speed loop. Compared with the QPR controller alone, the IP+QPR integrated controller also significantly improves the response speed, achieving the goal of optimized design.

[0091] Although the embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and cannot be understood as limiting the present invention. A person skilled in the art may change, modify, replace and modify the above embodiments within the scope of the present invention without departing from the principles and purpose of the present invention. The protection scope of the present invention is defined by the claims and their equivalent technical solutions.

Claims

1. A method for suppressing motor torque pulsation based on integral proportional and quasi-proportional resonant control, characterized in that: The following steps are involved: S1: The actual position of the motor rotor is obtained through sensor sampling to calculate its speed n; S2: Difference between the current speed and the target speed, and obtain the dq axis stator current i required for current control through the integral proportional and quasi-proportional resonant integrated controller. dref with i qref ; S3: i dref with i d 、i qref with i q Make a difference and get u through PI controller q with u d , and then after the inverse Park transformation, it is input into the space vector modulation module, and the PWM control signal is output to control the motor; where i d 、i q are the stator currents of the dq axis, u d 、u q are the stator voltages of the dq axes respectively.

2. The method for suppressing motor torque pulsation according to claim 1, characterized in that: In step S1, a sensor is used to sample the position of the rotor, and the rotor speed n is calculated by the position difference at each moment Δt.

3. The motor torque pulsation suppression method according to claim 1, characterized in that: In step S3, the three-phase current i is sampled and obtained. a 、i b 、i c , the sampled three-phase current is subjected to Clark transformation and Park change to obtain i q with i d .

4. The method for suppressing motor torque pulsation according to claim 1, characterized in that: In step S2, a parallel quasi-proportional resonant controller is used as part of the speed loop, and its transfer function is: Where K p is the proportionality coefficient, ω c is the cut-off angular frequency, ω is the resonant angular frequency, and K is the resonant parameter.

5. The motor torque pulsation suppression method according to claim 1, characterized in that: In step S2, the integral proportion and quasi-proportional resonant controllers are controlled in parallel, and the parameters of the two controllers are adjusted separately; the integral proportion controller is adjusted by zero-pole adjustment, and the bandwidth is selected from 1 / 10 to 1 / 5 of the current loop.

Images