阅读说明：本技术 一种内置式永磁同步电机无位置传感器参数误差补偿策略 (Built-in permanent magnet synchronous motor position sensorless parameter error compensation strategy ) 是由 张航 梁文睿 张辉 于 2021-06-03 设计创作，主要内容包括：本发明公开了一种内置式永磁同步电机无位置传感器参数误差补偿策略,具体为：基于磁链间接计算法,对MT轴电流进行估计并用给定磁链代替估计值,从而改进基于磁链的转子位置估计；获取偏差系数μ和λ,对永磁体磁链和交轴电感参数进行实时在线修正,并将修正后的参数用于电机的最大转矩电流比控制定子磁链给定值计算、MT轴电流估计以及负载角的计算。本发明针对低开关频率SHEPWM调制下的内置式永磁同步电机无位置传感器控制,提出了一种参数误差补偿策略,可以实时修正永磁体磁链和交轴电感参数偏差,从而提升转子位置估计精度,增强系统鲁棒性。(The invention discloses a built-in permanent magnet synchronous motor position sensorless parameter error compensation strategy, which specifically comprises the following steps: estimating the MT shaft current based on a flux linkage indirect calculation method, and replacing the estimated value with a given flux linkage, thereby improving the flux linkage-based rotor position estimation; and acquiring deviation coefficients mu and lambda, performing real-time online correction on the permanent magnet flux linkage and quadrature axis inductance parameters, and using the corrected parameters for calculating the maximum torque current ratio control stator flux linkage set value of the motor, estimating the MT axis current and calculating the load angle. The invention provides a parameter error compensation strategy aiming at the position-sensorless control of a built-in permanent magnet synchronous motor under the modulation of low switching frequency SHEPWM, and the parameter deviation of permanent magnet flux linkage and quadrature axis inductance can be corrected in real time, so that the estimation precision of the rotor position is improved, and the robustness of the system is enhanced.)

1. A built-in permanent magnet synchronous motor position sensorless parameter error compensation strategy is characterized by being implemented according to the following steps:

step 1, estimating the current of an MT shaft based on a flux linkage indirect calculation method, and replacing an estimated value with a given flux linkage, thereby improving the rotor position estimation based on the flux linkage;

and 2, acquiring deviation coefficients mu and lambda, performing real-time online correction on the permanent magnet flux linkage and quadrature axis inductance parameters, and using the corrected parameters for calculating the maximum torque current ratio control stator flux linkage given value of the motor, estimating the MT axis current and calculating the load angle.

2. The strategy for compensating the position sensorless parameter error of the interior permanent magnet synchronous motor according to claim 1, wherein in the step 1, specifically:

step 1.1, for three-phase current i_A、i_B、i_CSampling, performing Clarke transformation to obtain alpha-axis and beta-axis currents, and calculating stator magnetic chain angle estimation valueAnd an electromagnetic torque estimateAs shown in formula (1) and formula (2);

in the formula, #_αIs an alpha axis stator flux linkage; psi_βIs a beta axis stator flux linkage; p is a radical of_nIs the number of pole pairs; i.e. i_αIs the α axis current; i.e. i_βIs the beta axis current;

step 1.2, calculating the stator flux linkage given valueBased on the electromagnetic torque estimation value, as shown in equation (3)Calculating an M-axis estimated currentAnd T-axis estimated currentAs shown in formula (4) and formula (5); q-axis set current controlled by maximum torque current ratioWith d-axis set currentRepresented by formula (6) and formula (7);

in the formula, #_fIs a permanent magnet flux linkage; i.e. i_sIs the stator current; l is_dIs a d-axis inductance; l is_qIs a q-axis inductance; a ═ psi_fL_d/l；C＝1/l；Median valuep_nIs the number of pole pairs;

step 1.3, calculating the load angleBased on the stator flux angle estimate, as shown in equation (8)Calculating rotor position estimatesAs shown in formula (9);

in the formula, #_sA stator flux linkage;is the M-axis estimated current;is the T-axis estimated current;

step 1.4, calculating the compensated rotor positionAs shown in formula (10);

in the formula (I), the compound is shown in the specification,is the angular speed of the rotor, t_rUpdating the time for PWM; t is t_sThe most recent sample time before the update.

3. The strategy for compensating the position-sensorless parameter error of the interior permanent magnet synchronous motor according to claim 2, wherein in the step 2, the specific steps are as follows;

step 2.1, actual voltage u based on d axis_d-motAnd d-axis command voltage u_d-calCalculating q-axis inductance deviation coefficient mu and permanent magnet flux linkage deviation coefficient lambda as shown in formula (11) and formula (12);

in the formula (I), the compound is shown in the specification,is the rotor angular velocity; i.e. i_dIs the d-axis current; i.e. i_qIs the q-axis current;

step 2.2, calculating the actual value L of the compensated q-axis inductance_q-acAnd the actual value psi of the permanent magnet flux linkage after compensation_f-acAs shown in formula (13) and formula (14), respectively;

L_q-ac＝(1+μ)L_q (13)；

ψ_f-ac＝(1+λ)ψ_f (14)；

step 2.3, after obtaining the parameter deviation coefficients in the formula (11) and the formula (12), the permanent magnet flux linkage psi is processed according to the formula (13) and the formula (14)_fAnd q-axis inductance L_qPerforming real-time online correction by L_q-acIn place of L_qBy psi_{f_ac}Instead of psi_fModifying the stator flux linkage set value calculation, the MT axis current estimation, the load angle calculation and the permanent magnet flux linkage and q axis inductance parameters in the MTPA control, and then rewriting the formulas (3), (4), (5), (6), (7) and (8) as shown in the formulas (15), (16), (17), (18), (19) and (20);

step 2.4, d-axis and q-axis given currentsAndobtaining d-axis and q-axis given voltages after being adjusted by a current regulatorAndcalculating modulation degree M, voltage vector angle beta and compensated voltage vector angle beta_{re_com}The switching angle alpha can be obtained by the on-line table look-up method for the modulation degree M_NFromAnd beta_{re_com}Adding to obtain the voltage vector angle under the ABC coordinate systemTo pairAnd alpha_NAnd performing pulse reconstruction to obtain three-phase pulse output.

4. The PSSM position sensorless parameter error compensation strategy according to claim 3, wherein in step 2.4, the modulation M, the voltage vector angle β and the compensated voltage vector angle β_{re_com}The calculation formulas of (a) are respectively shown as a formula (21), a formula (22) and a formula (23);

in the formula of U_dcIs a dc voltage.

Technical Field

The invention belongs to the technical field of transmission control of alternating current motors, and particularly relates to a position-sensorless parameter error compensation strategy for a built-in permanent magnet synchronous motor.

Background

With the rapid development of high-speed rail trains, permanent magnet synchronous traction motors with wide speed regulation range, high power density and low energy consumption become research hotspots in the current rail traffic field, and the existing motor train unit train traction systems all adopt mechanical position/speed sensors to acquire position or rotating speed signals of the motors. In the actual running process of a high-speed train, the electromagnetic environment is complex, vibration is severe, failure of a mechanical sensor is easily caused, a traction system is caused to break down, large torque impact is caused, critical components such as a bearing, a gear and a motor are damaged in serious conditions, and the running safety of the train is damaged. The driving technology without the position sensor can fundamentally eliminate the potential safety hazard and has the advantages of strong anti-jamming capability, high integration level, long service cycle and the like.

For a high-speed rail traction system, the switching frequency of the IGBT is usually about 500Hz due to the limitation of switching loss and heat dissipation. In order to obtain better current and voltage performance in a full speed range, reduce switching loss, prolong the service life of a high-power switching tube, obtain good inverter voltage output characteristics and fully utilize bus voltage, asynchronous modulation is generally used at zero low speed, segmented synchronous modulation is used at medium and high speed, and square wave modulation is used above rated rotation speed.

When a modulation mode of a traction inverter is actually determined, a switching rotating speed is usually determined according to a carrier ratio, at present, an optimized synchronous modulation mode such as Specific Harmonic Elimination (SHEPWM) is usually used when the carrier ratio is smaller than 10, the running time of a corresponding mode accounts for about half of the running time of a train, and at the moment, the corresponding rotating speed range is usually estimated by using a back electromotive force or a flux linkage model excited by a fundamental frequency. Compared with a zero-low-speed high-frequency signal injection method based on saliency tracking, the method does not need to consider rotor saliency, does not need extra harmonic signal injection, is simple in digital implementation, and is mature in application in the industrial field. For the rotor position observer based on the motor model, the construction is based on a voltage, current or flux linkage equation, so the rotor position estimation precision and the system robustness are greatly determined by the accuracy of the command voltage, the feedback current and the motor parameters. Flux linkage based position sensorless control is typically combined with direct torque or stator field orientation control. At this time, the rotor position is usually calculated by a rotor or stator flux linkage vector, and mainly includes two ways: an "effective flux linkage" method based on an extended rotor flux linkage model and an indirect calculation method based on a load angle and a stator flux linkage angle. The rotor position estimation precision and the closed loop robustness of the two modes are determined by the precision of flux linkage amplitude and angle, and are easily influenced by flux linkage, inductance parameter change and system delay. Therefore, in the permanent magnet traction system, parameter variation and system delay characteristics under optimized synchronous modulation are combined, permanent magnet flux linkage and quadrature axis inductance parameter deviation are corrected, and the method has important practical significance for improving the control performance of a position-sensorless control system.

Disclosure of Invention

The invention aims to provide a position sensorless parameter error compensation strategy for a built-in permanent magnet synchronous motor, which corrects parameter error compensation of permanent magnet flux linkage and quadrature axis inductance in real time by solving a deviation coefficient, thereby improving the position estimation precision.

The technical scheme adopted by the invention is that a built-in permanent magnet synchronous motor position sensorless parameter error compensation strategy is implemented according to the following steps:

step 1, estimating the current of an MT shaft based on a flux linkage indirect calculation method, and replacing an estimated value with a given flux linkage, thereby improving the rotor position estimation based on the flux linkage;

and 2, acquiring deviation coefficients mu and lambda, performing real-time online correction on the permanent magnet flux linkage and quadrature axis inductance parameters, and using the corrected parameters for calculating the maximum torque current ratio control stator flux linkage given value of the motor, estimating the MT axis current and calculating the load angle.

The present invention is also characterized in that,

in the step 1, the method specifically comprises the following steps:

step 1.1, for three-phase current i_A、i_B、i_CSampling, performing Clarke transformation to obtain alpha-axis and beta-axis currents, and calculating stator magnetic chain angle estimation valueAnd an electromagnetic torque estimateAs shown in formula (1) and formula (2);

in the formula, #_αIs an alpha axis stator flux linkage; psi_βIs a beta axis stator flux linkage; p is a radical of_nIs the number of pole pairs; i.e. i_αIs the α axis current; i.e. i_βIs the beta axis current;

step 1.2, calculating the stator flux linkage given valueBased on the electromagnetic torque estimation value, as shown in equation (3)Calculating an M-axis estimated currentAnd T-axis estimated currentAs shown in formula (4) and formula (5); q-axis set current controlled by maximum torque current ratioWith d-axis set currentRepresented by formula (6) and formula (7);

in the formula, #_fIs a permanent magnet flux linkage; i.e. i_sIs the stator current; l is_dIs a d-axis inductance; l is_qIs a q-axis inductance; a ═ psi_fL_d/l；c＝1/l；Median valuep_nIs the number of pole pairs;

step 1.3, calculating the load angleBased on the stator flux angle estimate, as shown in equation (8)Calculating rotor position estimatesAs shown in formula (9);

in the formula, #_sA stator flux linkage;is the M-axis estimated current;is the T-axis estimated current;

step 1.4, calculating the compensated rotor positionAs shown in formula (10);

in the formula (I), the compound is shown in the specification,is the angular speed of the rotor, t_rUpdating the time for PWM; t is t_sThe most recent sample time before the update.

In the step 2, the concrete steps are as follows;

step 2.1, actual voltage u based on d axis_d-motAnd d-axis command voltage u_d-calCalculating q-axis inductance deviation coefficient mu and permanent magnet flux linkage deviation coefficient lambda as shown in formula (11) and formula (12);

in the formula (I), the compound is shown in the specification,is the rotor angular velocity; i.e. i_dIs the d-axis current; i.e. i_qIs the q-axis current;

step 2.2, calculating the actual value L of the compensated q-axis inductance_q-acAnd the actual value psi of the permanent magnet flux linkage after compensation_f-acAs shown in formula (13) and formula (14), respectively;

L_q-ac＝(1+μ)L_q (13)；

ψ_f-ac＝(1+λ)ψ_f (14)；

step 2.3, after obtaining the parameter deviation coefficients in the formula (11) and the formula (12), the permanent magnet flux linkage psi is processed according to the formula (13) and the formula (14)_fAnd q-axis inductance L_qPerforming real-time online correction by L_q-acIn place of L_qBy psi_f-acInstead of psi_fModifying the stator flux linkage set value calculation, the MT axis current estimation, the load angle calculation and the permanent magnet flux linkage and q axis inductance parameters in the MTPA control, and then rewriting the formulas (3), (4), (5), (6), (7) and (8) as shown in the formulas (15), (16), (17), (18), (19) and (20);

step 2.4, d-axis and q-axis given currentsAndobtaining d-axis and q-axis given voltages after being adjusted by a current regulatorAndcalculating modulation degree M, voltage vector angle beta and compensated voltage vector angle beta_re-comThe switching angle alpha can be obtained by the on-line table look-up method for the modulation degree M_NFromAnd beta_re-comAdding to obtain the voltage vector angle under the ABC coordinate systemTo pairAnd alpha_NAnd performing pulse reconstruction to obtain three-phase pulse output.

In step 2.4, the modulation M, the voltage vector angle β and the compensated voltage vector angle β_re-comThe calculation formulas of (a) are respectively shown as a formula (21), a formula (22) and a formula (23);

in the formula of U_dcIs a dc voltage.

The invention has the beneficial effects that:

1) the robustness improvement of the built-in permanent magnet synchronous motor position-sensorless control under the low switching frequency SHEPWM modulation is realized;

2) improving the traditional indirect calculation method, calculating a load angle by using the estimated dq axis current and replacing an estimated value by using a flux linkage set value;

3) the parameter error compensation of the permanent magnet flux linkage and the quadrature axis inductance is corrected in real time by solving the deviation coefficient, so that the position estimation precision is improved.

Drawings

FIG. 1 is a schematic block diagram of a position sensorless parameter error compensation strategy for an interior permanent magnet synchronous motor of the present invention;

FIG. 2 is a block diagram of a hardware circuit structure of an experimental system used in a position sensorless parameter error compensation strategy of an interior permanent magnet synchronous motor according to the present invention;

FIG. 3 is a waveform diagram of a current performance test of the present invention under SHE7 pulse-optimized synchronous modulation without using the parameter error compensation strategy;

FIG. 4 is a waveform diagram of a current performance test of the present invention under SHE7 pulse-optimized synchronous modulation using the parameter error compensation strategy;

FIG. 5 is a waveform diagram of the performance test of the speed and position estimation without using the parameter error compensation strategy under SHE7 pulse optimization synchronous modulation according to the present invention;

FIG. 6 is a waveform diagram of the performance test of the speed and position estimation using the parameter error compensation strategy under SHE7 pulse optimization synchronous modulation according to the present invention;

Detailed Description

The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

The invention discloses a built-in permanent magnet synchronous motor position sensorless parameter error compensation strategy, which is implemented according to the following steps:

step 1, estimating the MT axis current based on a flux linkage indirect calculation method, and replacing the estimated value with a given flux linkage, thereby improving the flux linkage-based rotor position estimation, and the schematic block diagram is shown in fig. 1, and the specific steps are as follows:

step 1.1, for three-phase current i_A、i_B、i_CSampling, performing Clarke transformation to obtain alpha-axis and beta-axis currents, and calculating stator magnetic chain angle estimation valueAnd an electromagnetic torque estimateAs shown in formula (1) and formula (2);

in the formula, #_αIs an alpha axis stator flux linkage; psi_βIs a beta axis stator flux linkage; p is a radical of_nIs the number of pole pairs; i.e. i_αIs the α axis current; i.e. i_βIs the beta axis current;

step 1.2, calculating the stator flux linkage given valueBased on the electromagnetic torque estimation value, as shown in equation (3)Calculating an M-axis estimated currentAnd T-axis estimated currentAs shown in formula (4) and formula (5); q-axis set current by maximum torque current ratio control (MTPA)With d-axis set currentRepresented by formula (6) and formula (7);

in the formula, #_fIs a permanent magnet flux linkage; i.e. i_sIs the stator current; l is_dIs a d-axis inductance; l is_qIs a q-axis inductance; a ═ psi_fL_d/l；C＝1/l；Median valuep_nIs the number of pole pairs;

step 1.3, calculating the load angleBased on the stator flux angle estimate, as shown in equation (8)Calculating rotor position estimatesAs shown in formula (9);

in the formula, #_sA stator flux linkage;is the M-axis estimated current;is the T-axis estimated current;

step 1.4, calculating the compensated rotor positionAs shown in formula (10);

in the formula，Is the angular speed of the rotor, t_rUpdating the time for PWM; t is t_sThe latest sampling moment before updating;

based onCarrying out Park conversion to obtain d-axis current i_dAnd q-axis current i_qThereby forming a current loop, pairDifferential calculation and simple digital low-pass filtering are adopted to carry out rotating speed estimation so as to form a rotating speed ring and control the motor;

step 2, after obtaining deviation coefficients mu and lambda in an internal permanent magnet synchronous motor driving system, carrying out real-time online correction on permanent magnet flux linkage and quadrature axis inductance parameters, and using the corrected parameters for maximum torque current ratio control (MTPA) of the motor, stator flux linkage set value calculation, MT axis current estimation and load angle calculation, wherein the specific steps are as follows;

step 2.1, actual voltage u based on d axis_d-motAnd d-axis command voltage u_d-calCalculating q-axis inductance deviation coefficient mu and permanent magnet flux linkage deviation coefficient lambda as shown in formula (11) and formula (12);

in the formula (I), the compound is shown in the specification,is the rotor angular velocity;is d-axis given current;is q-axis given current; i.e. i_dIs the d-axis current; i.e. i_qIs the q-axis current;

step 2.2, calculating the actual value L of the compensated q-axis inductance_q-acAnd the actual value psi of the permanent magnet flux linkage after compensation_f-acAs shown in formula (13) and formula (14), respectively;

L_q-ac＝(1+μ)L_q (13)；

ψ_f-ac＝(1+λ)ψ_f (14)；

step 2.3, the compensated parameters are used in motor control, that is, after obtaining the parameter deviation coefficients in the equations (11) and (12), the permanent magnet flux linkage psi of the corresponding module in fig. 1 can be mapped according to the relationship between the equations (13) and (14)_fAnd q-axis inductance L_qPerforming real-time online correction by L_q-acIn place of L_qBy psi_f-acInstead of psi_f. Modifying the given value calculation of stator flux linkage, the estimation of MT axis current, the calculation of load angle and the permanent magnet flux linkage and q axis inductance parameters in MTPA control, and rewriting the formulas (3), (4), (5), (6), (7) and (8) as shown in the formulas (15), (16), (17), (18), (19) and (20);

step 2.4, d-axis and q-axis given currentsAndobtaining d-axis and q-axis given voltages after being adjusted by a current regulatorAndthe modulation degree M is calculated from the equation (21), the voltage vector angle beta is calculated from the equation (22), and the compensated voltage vector angle beta is calculated_re-comAs shown in formula (23), the switching angle α can be obtained from the modulation M by an online table look-up method_NFromAnd beta_re-comAdding to obtain the voltage vector angle under the ABC coordinate systemTo pairAnd alpha_NAnd performing pulse reconstruction to obtain three-phase pulse output.

In the formula (I), the compound is shown in the specification,is the angular speed of the rotor, t_rUpdating the time for PWM; t is t_sThe latest sampling moment before updating;is the estimated electromagnetic torque; l is_dIs a d-axis inductor; p is a radical of_nIs the number of pole pairs;is the d-axis given voltage;is the q-axis given voltage; u shape_dcIs a direct current voltage;

the built-in permanent magnet synchronous motor is easily influenced by parameter changes of flux linkage and inductance, so that the following problems occur: errors of rotor flux linkage and quadrature axis inductance of the permanent magnet motor can affect position estimation accuracy and closed loop robustness. In order to improve the robustness of the built-in permanent magnet synchronous motor position sensorless control, the invention provides a parameter error compensation strategy aiming at the built-in permanent magnet synchronous motor position sensorless control under the modulation of low switching frequency SHEPWM, and the parameter deviation of permanent magnet flux linkage and quadrature axis inductance can be corrected in real time, so that the rotor position estimation precision is improved, and the system robustness is enhanced.

The system hardware structure of the present invention is shown in fig. 2, and includes: the system comprises a rectification circuit, a filter circuit, a three-phase full-bridge inverter, an IPMSM (interior permanent magnet synchronous motor), an FPGA controller, an isolation driving circuit, a rotary transformer and a current acquisition circuit; the system adopts a rotary transformer to collect real position signals and compares the real position signals with an estimated position. The output end of a three-phase full-bridge inverter in the control system is connected with an IPMSM stator three-phase winding, and the IPMSM is controlled after the initial position of a rotor is estimated. Fig. 3 to 6 are graphs showing the comparison between the current performance and the rotation speed and position estimation performance of the IPMSM under SHE7 pulse-optimized synchronous modulation under the control of the hardware system shown in fig. 2 and using the parameter compensation strategy, and the comparison without using the parameter compensation strategy. The current performance waveform without this parameter compensation strategy is shown in fig. 3: fundamental wave current obviously has phase lag and amplitude attenuation, the sine degree of current waveform is also deteriorated, magnetic field orientation error causes larger alternating-direct axis current fluctuation, and harmonic energy is mainly concentrated in 11 th harmonic, 13 th harmonic and 19 th harmonic in PSD distribution. The waveform diagram of the current performance after the parameter compensation strategy is used is shown in FIG. 4: the phase lag and amplitude error of the fundamental current are eliminated, the alternating-direct axis current fluctuation is controlled within 1A, the PSD distribution is more even, and the amplitude of each subharmonic is weakened. The speed and position estimation performance profiles without the parameter compensation strategy are shown in FIG. 5: the estimated rotating speed has larger fluctuation, the average fluctuation amplitude of the electrical angular speed can reach 300r/min, namely in a high rotating speed area, the parameter error can increase the current oscillation to cause the rotating speed fluctuation, the estimated rotor position is also distorted, and the maximum position estimation error exceeds 0.3 rad. The waveform diagram of the rotation speed and position estimation performance after the parameter compensation strategy is used is shown in FIG. 6: the fluctuation of the estimated signal is eliminated and the position estimation error does not exceed 0.1 rad.