Periodic update and cyclic assignment low speed servo system center current suppression method

Abstract

The application relates to a low-speed servo system center current suppression method of periodical updating and cyclic assignment. The method comprises the following steps: dividing p electric angle integral windows with a width of 2pi from the rotor zero position to a 2pi mechanical angle position after one rotation of the motor; simultaneously integrating three-phase currents in the same electric angle integral window, and respectively obtaining the direct current bias of each phase current under the magnetic pole pair in different electric angle integral windows; designing mp mean filtering modules according to the motor pole pair number p and the motor phase number m, and inputting the p integral results of each phase winding in one rotation of the motor into the mp mean filtering modules for filtering processing; and adopting a cyclic assignment method to update the center currents of each phase in real time, performing clark and park coordinate transformation on the phase currents after bias correction to obtain motor d and q axis currents after first harmonic suppression, so that dynamic tracking of the center currents of each phase of the permanent magnet synchronous motor is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of harmonic suppression of control systems, and particularly to a method for suppressing the center current of a low-speed servo system with periodic updating and cyclic assignment. BACKGROUND

[0002] Chip process defects and power supply ripple interference are common phenomena in permanent magnet synchronous motor low-speed servo systems (low speed in this paper refers to speed lower than 500 RPM), which will cause one harmonic pulsation of torque current. The existence of low-frequency harmonic current seriously affects the stability of low-speed servo system operation and the reliability of space missions.

[0003] In the field of low-speed control, permanent magnet synchronous motor (PMSM) direct-drive servo systems are widely used in aerospace due to their strong load-carrying capacity, fast dynamic response, high stability, and small size. Vector control uses coordinate transformation to decouple the excitation current and torque current of the permanent magnet synchronous motor, and controls the electromagnetic torque by controlling the current vector. As the speed of the servo system decreases, the rotational inertia filter effect decays, the frequency of the phase current decreases, and the influence of harmonic current becomes more and more obvious. When the speed is lower than a certain critical value, the fluctuation of electromagnetic torque will cause the system to appear low-speed jitter or low-speed crawling phenomenon, as described in the references "Xia X, Zhang B, Li X. High Precision Low-Speed Control for Permanent Magnet Synchronous Motor. Sensors (Basel) [J]. 2020 Mar 10; 20(5)" and "Zhu, C.; Zeng, Z.; Zhao, R. Torque ripple elimination based on inverter voltage drop compensation for a three-phase four-switch inverter-fed PMSM drive under low speeds. IET Power Electron. 2017, 10, 1430-1437." Therefore, suppressing harmonic current is the key to improving the performance of low-speed systems.

[0004] The harmonic current of permanent magnet synchronous motor at low speed can be divided into low-order harmonic of current frequency and its multiples and high-order harmonic of inverter switching frequency and its multiples. The servo system itself presents low-pass characteristics, and high-frequency disturbance will be greatly attenuated, so the high-order harmonic current caused by PWM chopping has little effect on system stability. The generation of motor low-order harmonic current is mainly caused by the following two reasons: (1) motor itself. The non-symmetry of motor winding, the non-linear friction of the system, and the air gap magnetic field distortion caused by the tooth slot effect of the machine body, the distribution of the stator winding, the structure of the rotor magnetic pole, etc.; (2) drive control. The precision limit of the detection device, the current sampling quantization error, the unevenness of the circuit impedance, and the drift of the device characteristics, reference“Li S, Sun L Z, Liu X Y, An Q T. Harmonic current suppression strategy for permanent magnet synchronous motor [J]. Transactions of China Electrotechnical Society, 2019, 34(S1): 87-96”.

[0005] For the above influencing factors, domestic and foreign scholars have carried out a lot of research work from two aspects of improving the structure of the motor and optimizing the control strategy of the motor. Various methods have certain effect on suppressing harmonic current and improving motor current waveform. The literature "Li Jiebao, Zhang Yuejin. Analysis and weakening method of torque ripple of permanent magnet brushless motor [J]. Motor and control application, 2011, 38(04): 6-12+36" changes the pole arc coefficient, the magnetization direction of the magnetic pole, adopts the method of stator skew slot and fractional slot, and adjusts the motor parameters to reduce the value of the pulsating torque. In order to ensure the high stability of the aerospace system, the speed stability can be improved by increasing the cost of motor manufacturing to optimize the structure of the motor, but due to the ideal motor model cannot be realized, the harmonic ripple caused by the defects of the motor body still exists. The literature "Liao Yong, Zhen Shuai, Liu Zhen, Yao Jun. Harmonic injection to suppress torque ripple of permanent magnet synchronous motor [J]. Proceedings of the Chinese society of electrical engineering, 2011, 31(21): 119-127" uses rotating PI control and digital low-pass filter to construct harmonic current extraction module, which has obvious effect on 5th and 7th harmonic suppression of permanent magnet synchronous motor. For low-speed servo system, the realization of digital low-pass filter with low cut-off frequency and high filter order is low, and this method is not suitable for harmonic ripple caused by direct current bias of motor phase current. The literature "FAN Mingdi, Lin Hui, Lv Shuai. A method to suppress periodic speed ripple of PMSM-DTC [J]. Journal of motor and control, 2013, 17(09): 73-78+85" calculates the harmonic components of current based on FFT and realizes harmonic compensation combined with PI controller. The calculation accuracy of this algorithm is affected by the number of sampling points and sampling frequency. In low-speed servo system, the measurement error of low-frequency harmonic signal and the size of system operation amount are difficult to balance. The literature "Ji Kehui, Shen Jianxin. Low-speed motor servo system using disturbance torque observer [J]. Proceedings of the Chinese society of electrical engineering, 2012, 32(15): 100-106+9" and "Qu, JZ, Zhang, CN. Deadbeat harmonic current control of permanent magnet synchronous machine drives for torque ripple reduction. IEEE journal of emerging and selected topics in power electronics [J], 2022, 6" uses disturbance observer to suppress current harmonic, and the control effect is good, but the algorithm needs to set many parameters and has high complexity.

[0006] In the traditional method, the selection of the integral window is irrelevant to the phase, and the integral result of the last period is directly used for the bias correction of the next period. Due to the increase of the harmonic content of the winding magnetic motive force caused by the fractional-slot winding structure of the motor and the influence of the asymmetry of the magnetic circuit structure of each pole of the motor, the actual phase current of the motor is a signal with a period of 2nπ, which is composed of n periods of 2π, different amplitudes and non-ideal sine signals. Considering the actual application background of the non-ideal sine wave, the traditional control method cannot suppress the pulse noise of the center current, and does not consider the calculation error of the center current caused by the asymmetry of the magnetic circuit. SUMMARY

[0007] The purpose of the present application is to propose a periodic update and cyclic assignment low-speed servo system center current suppression method for realizing the suppression of the first harmonic ripple of torque.

[0008] In order to achieve the above purpose, the present application realizes the technical scheme as follows.

[0009] The present application proposes a periodic update and cyclic assignment low-speed servo system center current suppression method, which comprises:

[0010] According to the number of magnetic pole pairs of the motor, the electrical angle integral window is divided;

[0011] The three-phase current is integrated in the electrical angle integral window to obtain the direct current bias of the three-phase current under each magnetic pole pair of the motor;

[0012] The direct current bias of the three-phase current is subjected to mean filtering processing;

[0013] The three-phase current subjected to filtering processing is updated by using the cyclic assignment method;

[0014] The updated three-phase current is subjected to coordinate conversion;

[0015] The converted current is input into the current loop controller of the space low-speed servo system to realize the suppression of the center current of the low-speed servo system.

[0016] As one of the improvements of the above technical scheme, the electrical angle integral window is divided according to the number of magnetic pole pairs of the motor, and specifically comprises:

[0017] From the rotor zero position to the 2π mechanical angle position after one rotation of the motor, p electrical angle integral windows with a width of 2π are divided according to the number of magnetic pole pairs p of the motor.

[0018] As one of the improvements of the above technical scheme, the three-phase current is integrated in the electrical angle integral window to obtain the direct current bias of the three-phase current under each magnetic pole pair of the motor, and specifically comprises:

[0019] Integrate the three-phase currents simultaneously within the same electrical angle integral window, and obtain the DC bias of each phase current under each magnetic pole pair in different electrical angle integral windows.

[0020] As one of the improvements of the above technical solutions, the calculation formula of the three-phase current is:

[0021]

[0022] wherein, I A , I B , I C are three-phase currents respectively; I1, I2, I3 are DC biases of the three-phase currents respectively; I is a stator fundamental current peak value; ω is a rotor electrical angular velocity; t is time; is an initial phase of the fundamental current.

[0023] As one of the improvements of the above technical solutions, the DC bias of each phase current under each magnetic pole pair is obtained in different electrical angle integral windows while the three-phase currents are integrated simultaneously within the same electrical angle integral window, and the calculation formula is:

[0024]

[0025] wherein, θ is a rotor electrical angle.

[0026] As one of the improvements of the above technical solutions, the DC bias of the three-phase current is subjected to mean value filtering processing, and specifically includes:

[0027] According to the motor pole pair number p and the motor phase number m, mp mean value filtering modules are designed, and p integral results of each phase winding in one rotation of the motor are respectively input into each mean value filtering module for filtering processing.

[0028] As one of the improvements of the above technical solutions, the three-phase current subjected to filtering processing is updated by using a cyclic assignment method, and specifically includes:

[0029] The central currents under different magnetic pole pairs after filtering are sequentially stored in a register array with a width of p;

[0030] When the first data in the array is valid, the next moment current bias stored in the array is cyclically output based on the first-in first-out principle with 2π mechanical angle as a period, and the central current is updated.

[0031] As one of the improvements of the above technical solutions, the coordinate conversion is clark and park coordinate transformation.

[0032] Compared with the prior art, the present application has the following advantages:

[0033] The application optimizes the input signal of the current loop in the double closed-loop control system by adopting a low-speed servo system center current suppression method with periodic updating and cyclic assignment, combines the segmented integral window related to the pole pair number of the motor with the mean value filtering window, improves the harmonic suppression effect under the magnetic circuit asymmetry, and realizes the suppression of the primary low-frequency torque ripple.

[0034] (1) The mean value filtering module is introduced, which can effectively suppress the pulse noise of the center current.

[0035] (2) The application reduces the influence of the magnetic circuit asymmetry on the center current calculation result from two aspects:

[0036] ① The electrical angle segmented integral window is proposed; the 2π mechanical angle of one rotation of the motor is divided into p integral subintervals with 2π electrical angle as the window width; the integral intervals are independent of each other, and the integral results under each integral interval are independent of each other, which greatly reduces the calculation error caused by the magnetic circuit asymmetry;

[0037] ② The cyclic assignment method of the center current is adopted; compared with the continuous assignment method in which the current value of the current window is directly used for the next period, the cyclic assignment method uses the center current under a pair of magnetic poles in one rotation of the motor to correct the current offset under the same pair of magnetic poles in the next rotation, realizes the one-to-one correspondence between the center current and the rotor position, and further ensures the accuracy of the direct current offset calculation in each electrical angle period. BRIEF DESCRIPTION OF DRAWINGS

[0038] Figure 1 The structure diagram of the low-speed servo system center current suppression method with periodic updating and cyclic assignment;

[0039] Fig. 2(a) is the phase current before the center current algorithm is added, and Fig. 2(b) is the phase current after the center current algorithm is added;

[0040] Fig. 3(a) is the speed before the center current algorithm is added, and Fig. 3(b) is the speed after the center current algorithm is added;

[0041] Fig. 4(a) is the speed FFT before the current loop algorithm is added, and Fig. 4(b) is the speed FFT after the current loop algorithm is added;

[0042] Fig. 5(a) is a low-speed PMSM low-frequency harmonic current suppression system block diagram including the low-speed servo system center current suppression method with periodic updating and cyclic assignment of the application; and Fig. 5(b) is a low-speed PMSM low-frequency harmonic current suppression system block diagram including the parallel double QPR controller and the low-speed servo system center current suppression method with periodic updating and cyclic assignment of the application;

[0043] Figure 6The A-phase current fitting graph before and after the rotor rotates 180°;

[0044] Fig. 7(a) is the phase voltage before adding the center current algorithm, and Fig. 7(b) is the phase voltage after adding the center current algorithm;

[0045] Fig. 8(a) is the speed before and after adding the center current algorithm, and Fig. 8(b) is the speed FFT before and after adding the center current algorithm; DETAILED DESCRIPTION

[0046] The center current suppression method of a low-speed servo system based on periodic updating and cyclic assignment, that is, a spatial low-speed servo system control method based on cyclic assignment updating of the center current, comprises the following steps:

[0047] From the zero position of the rotor to the 2π mechanical angle position after the motor rotates one circle, p electrical angle integral windows with a width of 2π are divided;

[0048] The three-phase currents are simultaneously integrated in the same electrical angle integral window, and the DC bias of each phase current under the magnetic pole pair is obtained in different electrical angle integral windows;

[0049] According to the number of motor pole pairs p and the number of motor phases m, mp mean filtering modules are designed, and the p integral results of each phase winding in one rotation of the motor are respectively input into the mp mean filtering modules for filtering processing;

[0050] The center currents of each phase are updated in real time by using the cyclic assignment method, the phase currents after bias correction are subjected to clark and park coordinate transformation to obtain the motor d and q axis currents after first harmonic suppression, and dynamic tracking of the center currents of each phase of the permanent magnet synchronous motor is realized.

[0051] Based on the application background of aerospace low-speed servo systems, aiming at the low-frequency harmonic components that can greatly affect the low-speed stability of permanent magnet synchronous motors, the causes of the low-frequency harmonic components are analyzed, a center current suppression method of a low-speed servo system based on periodic updating and cyclic assignment is proposed for the first harmonic ripple of torque, and finally the effectiveness and feasibility of the control strategy are verified through the simulation model and experimental results built on the Matlab / Simulink platform.

[0052] 1. Analysis of low-frequency harmonic currents

[0053] 1.1 First harmonic

[0054] In permanent magnet synchronous motor servo control system, the current detection circuit is often built by Hall current sensor chip. For the convenience of analysis, it is assumed that the three-phase stator winding parameters are symmetrical, the core saturation is not considered, and the influence of hysteresis and eddy current loss is ignored. The current is affected by many factors and superimposed with DC bias in the process of sensor, low-pass filter to AD conversion circuit. For example, due to the chip manufacturing process defects, the three sensors will not correspond to the same reference voltage at 0A, which will also cause the zero error of AD conversion chip; when the power signal and ground wire exist ripple interference, the output of operational amplifier and AD conversion chip will produce DC bias. In abc three-phase stationary coordinate system, the PMSM current equation with DC bias can be written as

[0055]

[0056] In the formula, I A , I B , I C are the A, B, C axis components of stator current respectively; I1, I2, I3 are the DC bias of A, B, C axis respectively; I is the peak value of stator fundamental current; ωt is the rotor electrical angle; is the initial phase of fundamental current. The three-phase current is constrained by constant amplitude, and after clark and park coordinate transformation, the current equation of PMSM in dq synchronous rotating coordinate system with q axis leading d axis by 90° is

[0057]

[0058]

[0059]

[0060]

[0061] In the formula, I d , I q are the d, q axis components of stator current respectively; I d1 , I q1 are the DC components of stator fundamental in d, q axis respectively; is the first harmonic peak value of DC bias in dq coordinate system; γ is the electrical angle fluctuation of first harmonic; a, b are the DC components of DC bias in α, β axis respectively. In the vector control mode of I d = 0, considering the first harmonic in q axis, the motor electromagnetic torque can be expressed as:

[0062]

[0063] In the formula, T e is the motor electromagnetic torque; P is the electromagnetic power; ω m ​is the mechanical angular velocity; p is the number of motor pole pairs; is the rotor permanent magnet flux linkage. From the above equation, when the motor phase current exists DC bias, the first harmonic component of q-axis current will cause the first harmonic ripple of torque.

[0064] 2 Current harmonic suppression strategy

[0065] Based on the different factors of each harmonic current and the nonlinear and strong coupling characteristics of PMSM, if the above harmonics are compensated as a whole, the suppression effect of each harmonic will not be ideal. To optimize the performance of low-speed servo system, this paper designs an algorithm from the causes of harmonic current, and gradually weakens each harmonic component that has a greater impact on torque current.

[0066] 2.1 Periodic update and cyclic assignment of low-speed servo system center current suppression method

[0067] When the system is working, the three-phase center current with the same size and constant will inevitably introduce DC bias due to the influence of factors such as chip zero error and circuit ripple interference. To suppress the first harmonic ripple of torque, from equation (1-1), a window with a width of 2π electrical angle is set, and the three-phase current is integrated in the window range to obtain the DC bias of each phase current in the current period. Due to the influence of the asymmetry of the motor magnetic circuit, the number of stator teeth and slots under adjacent magnetic poles is different, and the current waveform of the motor rotating one circle at any time is not completely consistent. The relationship between the number of slots per phase per pole q and the number of teeth Z is

[0068]

[0069] In the formula, p is the number of pole pairs; m is the number of phases. According to whether q is an integer or a fraction, PMSM can be divided into two categories: integer slot and fractional slot. Fractional slot motor weakens the tooth harmonic electromotive force induced by the tooth effect, and is more suitable for low-speed servo control. At the same time, fractional slot winding increases the harmonic content of the air gap magnetic field, causing increased rotor iron loss, increased unbalanced magnetic pull of the machine body, and increased vibration and noise. Considering the application background of low-speed system, this paper selects fractional slot motor as the controlled object. Due to the more prominent difference of the waveform under different electrical angle windows in one rotation of fractional slot motor, in order to realize effective suppression of the DC bias of phase current, the electrical angle sliding integral window is converted into a segmented integral window under the number of motor pole pairs.

[0070]

[0071] where θ is the rotor electrical angle. In the process of one revolution of the motor, 2π mechanical angle is divided into p integral windows according to the motor pole pair number p, and the integral values I1, I2, I3 of each section are calculated in turn. After mean filtering processing, the center current of each phase is updated in different window intervals by using the cyclic assignment method, so as to realize the dynamic tracking of the center current of the permanent magnet synchronous motor. Taking the permanent magnet synchronous motor with a pole pair number of 2 and a mean filter with a window width of 2 as an example, the structure of the low-speed servo system center current suppression method based on periodic update and cyclic assignment is shown in Fig. 3. Figure 1

[0072] 3 Simulation and experimental result analysis

[0073] 3.1 Simulation result analysis

[0074] Based on the low-speed servo system center current suppression method based on periodic update and cyclic assignment proposed in this paper, a system simulation model is established in the Matlab / Simulink environment. The current harmonics and torque ripple before and after the algorithm are analyzed to verify the effectiveness of the algorithm. Since the Simulink simulation model does not consider the nonlinear factors such as air gap magnetic field distortion and motor slot effect, this paper introduces the 1st harmonic of the phase current by adjusting the direct current bias of the phase current. The simulation parameters of the space vector pulse width modulation system (SVPWM) are shown in Table 3-1, and the simulation parameters of the permanent magnet synchronous motor are shown in Table 3-2.

[0075] Table 3-1 SVPWM simulation parameters

[0076] DC bus voltage carrier period sampling time dead time U dc / V]] T / μs T S / μs]]> T d / μs]]> 12 200 0.2 0.2

[0077] Table 3-2 Permanent magnet synchronous motor simulation parameters

[0078] load torque moment of inertia phase inductance phase resistance pole pairs T e / N·m]]> J / kg-m 2 ]] L / mH R / Ω p / pairs 0 1.48 x 10 -5 ]]> 2.3 1.66 1

[0079] The motor runs at no load at a speed of 5 rad / s, and the peak value of the phase current is about 0.58 A. At this time, the frequency of the fundamental current is about 0.796 Hz. Equation (3-1) is defined as the curve ripple parameter, which is used to measure the speed stability.

[0080]

[0081] where ω max is the maximum speed, and ω min is the minimum speed.

[0082] ​As shown in Fig. 2(a), it is the phase current before adding the center current algorithm, as shown in Fig. 2(b), it is the phase current after adding the center current algorithm; specifically, Fig. 2(a) is the phase current waveform when -0.1A, 0.15A, 0.2A DC bias exists in A, B, C phases respectively, after the current loop adds the harmonic suppression algorithm, as shown in Fig. 2(b), the current sinusoidal degree is obviously improved, and the total harmonic distortion (THD) of the system is reduced from 16.36% to 1.26%. As shown in Fig. 3(a), it is the rotating speed before adding the center current algorithm, as shown in Fig. 3(b), it is the rotating speed after adding the center current algorithm; by comparing Fig. 3(a) and Fig. 3(b), it can be seen that the DC bias can cause the increase of the first harmonic component of the rotating speed, and after adopting the center current suppression method of the low-speed servo system with periodic update and cyclic assignment, the rotating speed ripple η is reduced from 0.234% to 0.012%, thereby improving the system stability. The rotating speed is subjected to fast Fourier transform (FFT), as shown in Fig. 4(a) and Fig. 4(b), wherein Fig. 4(a) is the rotating speed FFT before adding the current loop algorithm, and Fig. 4(b) is the rotating speed FFT after adding the current loop algorithm, by comparison, it can be seen that the first harmonic content is reduced from 0.20241% to 0.00022%. It can be seen that the algorithm can effectively suppress the torque ripple caused by the DC bias.

[0083] 3.2 Analysis of experimental results

[0084] In order to further verify the practicability of the harmonic current suppression algorithm proposed in the present application, a permanent magnet synchronous motor drive control platform is built for experimental test. The motor rated parameters are: rated voltage U = 12V, tooth slot number n = 30, pole pair number p = 2, phase number m = 3. The experimental system block diagram is shown in Fig. 5(a) and Fig. 5(b), wherein Fig. 5(a) is a low-speed PMSM low-frequency harmonic current suppression system block diagram including the center current suppression method of the low-speed servo system with periodic update and cyclic assignment of the present application; Fig. 5(b) is a low-speed PMSM low-frequency harmonic current suppression system block diagram including the parallel double QPR controller and the center current suppression method of the low-speed servo system with periodic update and cyclic assignment of the present application. The system controller selects ZYNQ7015 of Xilinx Company, which combines the advantages of fast parallel processing speed of FPGA and low programming complexity of ARM floating point operation, so the control algorithm is mainly realized in two parts. The PL end is responsible for SVPWM module calculation control, inverter switch signal output and sensor data analysis and sending. The PS end receives the measured data, which is subjected to a harmonic suppression module and a coordinate transformation module, and outputs the voltage value of the motor in the αβ coordinate system, and realizes the data interaction between the FPGA through the high-speed AXI4 bus.

[0085] System test conditions are set to 52.08 rpm, load torque 0N, current fundamental frequency 1.738 Hz. The motor used in the experiment has a slot number per pole per phase q = 2.5, which is a fractional slot motor, that is, the slot numbers under each pair of magnetic poles are different, and the latter pair of poles is not necessarily a repetition of the former pair. Using Matlab / Cftool tool, the A-phase current in one rotation of the motor is Fourier fitted, Figure 6 The A-phase current comparison chart of the first 180° and the last 180° mechanical angle of the rotor after fitting is shown in the figure. It can be seen from the figure that due to the fractional slot characteristics, the phase currents of the two stages are not completely coincident, which verifies the effectiveness of the electrical angle segmentation integral window and the cyclic assignment method in the speed harmonic suppression algorithm.

[0086] As shown in Fig. 7(a), it is the phase voltage before adding the center current algorithm. Before adding the harmonic current suppression algorithm, the harmonic component of the motor phase voltage waveform is heavy, and the waveform distortion is serious. As shown in Fig. 7(b), it is the phase voltage after adding the center current algorithm. After adding the harmonic current suppression algorithm, the motor current waveform is obviously improved, and the sinusoidal degree is obviously improved. The B-phase voltage THD is reduced from 81.066% to 27.227%. As shown in Fig. 8(a), it is a comparison chart of the speed waveform before and after adding the periodic update and cyclic assignment low-speed servo system center current suppression method. Before adding the algorithm, the speed presents the pulsation of the fundamental frequency of the phase current. After adding the algorithm, the speed waveform is improved significantly, the speed variance is reduced from 1.368 to 0.5189, and the speed pulsation is reduced from 5.41% to 3.27%. The speed signal is subjected to FFT transformation, and Fig. 8(b) is a comparison chart of the speed after FFT transformation before and after adding the center current algorithm, wherein the signal frequency marked by the dotted line is 1.738 Hz. After adding the algorithm, the first harmonic content of the speed is reduced from 12.76% to 0.52%.

[0087] 4 Conclusion

[0088] In this paper, starting from the generation mechanism of the speed pulsation of the low-speed servo system, the low-frequency harmonic current is theoretically analyzed by using the mathematical model of the permanent magnet synchronous motor. A periodic update and cyclic assignment low-speed servo system center current suppression method is proposed for the first low-frequency torque pulsation, and the input signal of the current loop in the double closed-loop control system is optimized. Based on the above analysis and simulation experiment results, the following conclusions are obtained: the process defects of the current detection device, the ripple interference of the system detection circuit and other factors will cause the motor phase current to superimpose the direct current bias, causing the first harmonic jitter of the speed;

[0089] The above phenomenon is common in servo systems, even if the armature core has no tooth slot motor system, the first harmonic pulsation of the speed is still difficult to avoid. For the particularity of space application environment and task requirements, without increasing the hardware overhead, the design of high reliability control algorithm, the improvement of system anti-disturbance, the development trend of miniaturization, intelligence and integration of space servo system, the harmonic suppression algorithm under the high performance controller proposed in this paper has good practicability and foresight.

[0090] From the above specific description of the present application, for the first harmonic caused by the superposition of phase current and DC bias, a new low-speed servo system center current suppression method of periodic update and cyclic assignment is proposed. The segmented integral window related to the number of motor pole pairs is combined with the mean filter window to improve the harmonic suppression effect under the asymmetry of the magnetic circuit. This paper explains the mechanism of the harmonic suppression algorithm from the perspective of motor model and transfer function. The simulation and experiment show that the above algorithm has significant advantages in operation amount, implementation method, hardware resource overhead and harmonic suppression effect, which puts forward a new control direction for space low-speed servo system.

[0091] Finally, it should be noted that the above examples are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the examples, those skilled in the art should understand that modifications or equivalent replacements of the technical solutions of the present application do not deviate from the spirit and scope of the present application, and they should be covered in the scope of the claims of the present application.

Claims

1. A method for center current suppression of a low-speed servo system with periodic update and cyclic assignment, the method comprising: dividing an electrical angle integral window according to the number of motor magnetic pole pairs; wherein, from the rotor zero position to the 2π mechanical angle position after one rotation of the motor, p electrical angle integral windows with a width of 2π are divided according to the number of motor pole pairs p; simultaneously integrating three-phase currents in the same electrical angle integral window, and separately obtaining the DC bias of each phase current under each magnetic pole pair in different electrical angle integral windows; performing mean filtering processing on the DC bias of the three-phase currents; updating the three-phase currents after filtering processing using a cyclic assignment method, including: sequentially storing the center currents under different magnetic pole pairs after filtering into a register array with a width of p; when the first data in the array is valid, based on the first-in first-out principle, the current bias stored in the array at the next moment is output in a mechanical angle cycle to update the center currents of each phase; performing coordinate conversion on the updated three-phase currents; inputting the converted currents into a current loop controller of a spatial low-speed servo system to achieve center current suppression of the low-speed servo system.

2. The method of claim 1, wherein the low speed servo system center current suppression method is periodically updated and cyclically assigned. The spatial low-speed servo system is a double closed-loop servo system; the current loop and the speed loop of the double closed-loop servo system are both PI controllers.

3. The method of claim 1, wherein the low speed servo system center current suppression method is periodically updated and cyclically assigned. The calculation formula of the three-phase currents is: wherein I A , I B , I C are the three-phase currents, respectively; I1, I2, I3 are the DC offsets of the three-phase currents, respectively; I is the peak value of the stator fundamental current; ω is the rotor electrical angular velocity; t is time; is the initial phase of the fundamental current.

4. The method of claim 3, wherein the periodic update and circular assignment of low speed servo system center current is performed by a microprocessor. The calculation formula for simultaneously integrating three-phase currents in the same electrical angle integral window and separately obtaining the DC bias of each phase current under each magnetic pole pair in different electrical angle integral windows is: wherein, θ is the rotor electrical angle.

5. The method of claim 1, wherein the low speed servo system center current suppression method is periodically updated and cyclically assigned. The mean filtering processing on the DC bias of the three-phase currents specifically includes: designing mp mean filtering modules according to the number of motor pole pairs p and the number of motor phases m, and inputting the p integral results of each phase winding in one rotation of the motor into each mean filtering module for filtering processing.

6. The method of claim 1, wherein the low speed servo system center current suppression is periodically updated and cyclically assigned. The coordinate conversion is clark and park coordinate transformation.

Images