阅读说明：本技术 一种无权重系数的双电机驱动系统预测转矩控制方法 (Dual-motor drive system predicted torque control method without weight coefficient ) 是由 周扬忠 潘斌 陈艳慧 钟天云 于 2020-06-04 设计创作，主要内容包括：本发明涉及一种无权重系数的双电机驱动系统预测转矩控制方法。提供一六相串联三相双PMSM驱动系统,根据系统采样得到的包括六相电流、直流母线电压、位置角的信息,通过PI调节器得到控制两台PMSM在静止坐标系下的期望电压以及零序电压期望值,进而经<I>T</I><Sub>6</Sub>矩阵的逆变换得到六相相电压的期望值,将64个开关状态代入到基于六相相电压误差的成本函数中,令成本函数值最小的开关状态作用于系统下个周期。本发明实现了两台PMSM转矩和定子磁链幅值精准控制和零序电流的有效抑制；省去了烦琐的权重系数整定工作；并大大减少了预测转矩控制算法的计算量。(The invention relates to a control method for predicting torque of a dual-motor driving system without a weight coefficient. Providing a six-phase series three-phase double PMSM driving system, obtaining expected voltage and zero sequence voltage expected values of two PMSMs under a static coordinate system through a PI regulator according to information including six-phase current, direct current bus voltage and position angle obtained by system sampling, and further obtaining the expected voltage and zero sequence voltage expected values through the PI regulator T 6 The inverse transformation of the matrix obtains the expected value of the six-phase voltage, and the 64 switch states are substituted into the cost function based on the six-phase voltage error, so that the switch state with the minimum cost function valueThe next cycle of the system is acted on. The invention realizes the accurate control of the torque and the stator flux linkage amplitude of the two PMSM and the effective inhibition of the zero sequence current; the complicated weight coefficient setting work is saved; and greatly reduces the amount of calculation of the predicted torque control algorithm.)

1. A double-motor driving system prediction torque control method without weight coefficients is characterized in that a six-phase series three-phase double-PMSM driving system is provided, according to information including six-phase current, direct-current bus voltage and position angle obtained by system sampling, expected voltage and zero-sequence voltage expected values of two PMSMs under a static coordinate system are obtained through a PI regulator, and then the expected voltage and the zero-sequence voltage expected values are subjected to T₆And the inverse transformation of the matrix obtains the expected value of the six-phase voltage, and the 64 switching states are substituted into a cost function based on the six-phase voltage error, so that the switching state with the minimum cost function value acts on the next period of the system.

2. The method for controlling the predicted torque of the dual-motor drive system without the weight coefficient as claimed in claim 1 is implemented as follows:

step S1, using constant power transformation matrix T₆The six-phase current i obtained by sampling_A～i_FConverted into current i on α 1 β 1, α 2 β 2 and o1o2 coordinate systems_α1、i_β1、i_α2、i_β2、i_o1、i_o2：

Wherein i_α1、i_β1、i_α2、i_β2、i_o1、i_o2Current on α 1, β 1, α 2, β 2, o1 and o2 axes respectively, i_o1、i_o2For two zero sequence currents, i is because the neutral point of the three-phase PMSM is not led out_o1Is always 0;

step S2, obtaining stator flux psi of the two PMSMs on the static coordinate system according to the stator flux current model or the stator flux voltage model_sα1、ψ_sβ1、ψ_sα2、ψ_sβ2；ψ_sα1、ψ_sβ1、ψ_sα2、ψ_sβ2Stator flux linkages on α 1, β 1, α 2, β 2 axes, respectively;

1) if a current model of the stator flux linkage is used, the stator flux linkage psi_sα1ψ_sβ1、ψ_sα2ψ_sβ2Comprises the following steps:

wherein psi_f1、ψ_f2Permanent magnet flux linkage for two PMSM; l is_sσ1Is the self leakage inductance, L, of a six-phase PMSM phase winding_sm1＝(L_dm1+L_qm1)/2，L_rs1＝(L_dm1-L_qm1)/2，L_dm1、L_qm1The permanent magnet synchronous motor is characterized by comprising six-phase PMSM phase windings, a main magnetic flux direct shaft inductor and a quadrature axis inductor respectively; l is_sσ2Is the self leakage inductance, L, of the three-phase PMSM phase winding_sm2＝(L_dm2+L_qm2)/2，L_rs2＝(L_dm2-L_qm2)/2，L_dm2、L_qm2For main flux direct-axis and quadrature-axis inductances, theta, of three-phase PMSM phase windings_r1Is the angle between the d1 axis and the α 1 axis, theta_r2Is the angle between the d2 axis and the α 2 axis;

2) if a voltage model of the stator flux linkage is used, the stator flux linkage psi_sα1ψ_sβ1、ψ_sα2ψ_sβ2Comprises the following steps:

step S3, obtaining two PMSM torques T according to the formula (6) and the formula (7)_e1、T_e2：

T_e1＝p₁(ψ_sα1i_β1-ψ_sβ1i_α1)(6)

T_e2＝p₂(ψ_sα2i_β2-ψ_sβ2i_α2)(7)

Wherein p is₁Is six-phase PMSM pole pair number, p₂The number of pole pairs of the three-phase PMSM is shown; t is_e1For six-phase PMSM torque, T_e2Three-phase PMSM torque;

step S4, obtaining two PMSM stator flux linkage amplitude psi according to the formula (8) and the formula (9)_s1、ψ_s2：

Wherein psi_s1The six-phase PMSM stator flux linkage amplitude value; psi_s2The amplitude of the flux linkage of the three-phase PMSM stator is obtained;

step S5, according to the formula (10) and the formula (11), the given torque T of the two PMSMs is obtained by the PI regulators controlling the rotating speeds of the two PMSMs^* _e1、T^* _e2(ii) a Given value psi of six-phase PMSM stator flux linkage amplitude^* _s1Is composed ofGiven value psi of three-phase PMSM stator flux linkage amplitude^* _s2Is composed of

Wherein, ω is_r1、ω_r1A given electrical angular velocity and an actual electrical angular velocity for the six-phase PMSM; omega_r2、ω_r2A given electrical angular velocity and an actual electrical angular velocity for the six-phase PMSM; k_vp1、K_vi1Proportional coefficient and integral coefficient for controlling PI regulator of six-phase PMSM; k_vp2、K_vi2Proportional coefficients and integral coefficients for controlling a PI regulator of the three-phase PMSM;

step S6, obtaining the torque and stator flux linkage amplitude error delta T of the two PMSMs according to the formulas (12) to (15)_e1、Δψ_s1、ΔT_e2、Δψ_s2：

Step S7, adopting fixed parameter PI controller or variable ratio coefficient PI controller to calculate the torque angle variation delta needed for accurately eliminating the two PMSM torque and stator flux linkage amplitude errors₁、Δ₂；

1) If a fixed parameter PI controller is used, the amount of change in torque angle Δ is based on equations (16) and (17)₁、Δ₂Comprises the following steps:

Δ₁＝K_p1ΔT_e1+K_i1∫ΔT_e1dt(16)

Δ₂＝K_p2ΔT_e2+K_i2∫ΔT_e2dt(17)

wherein, K_p1、K_i1Proportional coefficient and integral coefficient of PI regulator for controlling six-phase PMSM torque; k_p2、K_i2Proportional coefficient and integral coefficient of PI regulator for controlling three-phase PMSM torque;

2) if a variable scale coefficient PI controller is adopted, the calculation model is as follows:

2.1) obtaining the scaling factor coefficients of the two PMSMs according to the formula (18) to the formula (21)

Wherein the content of the first and second substances,₁is a six-phase PMSM torque angle,₂is a three-phase PMSM torque angle; l is_d1＝L_sσ1+3L_sm1+3L_rs1Is a six-phase PMSM planar d-axis inductor, L_q1＝L_sσ1+3L_sm1-3L_rs1A six-phase PMSM plane q-axis inductor; l is_d2＝L_sσ1+2L_sσ2+3L_sm2+3L_rs2Is a three-phase PMSM planar d-axis inductor, L_q2＝L_sσ1+2L_sσ2+3L_sm2-3L_rs2A three-phase PMSM plane q-axis inductor;

2.2) obtaining the torque angle variation delta required for accurately eliminating the two PMSM torques and the stator flux linkage amplitude error according to the formula (22) and the formula (23)₁、Δ₂：

2.3) simultaneously, an integral link of the torque error adopts integral separation, and the integral separation only acts when the torque error is small and is used for compensating the inaccuracy of the parameters of the two PMSM;

step S8, obtaining expected values of stator flux linkage variation in α 1 β 1 and α 2 β 2 coordinate systems according to the formula (24) to the formula (27)

Wherein the content of the first and second substances,the expected value of the variation of the stator flux linkage under the α 1 β 1 coordinate system is obtained;is the expected value of the variation of the stator flux linkage in the α 2 β 2 coordinate system, T_sIs a control period;

step S9, obtaining expected values of the stator voltage under the coordinate system α 1 β 1 and α 2 β 2 according to the formula (28) and the formula (29)

Wherein the content of the first and second substances,the expected value of the stator voltage under the coordinate system of α 1 β 1;the expected value of the stator voltage under a α 2 β 2 coordinate system;

step S10, obtaining a zero sequence voltage given value u according to a formula (30)^* _o2：

Wherein, K_ip、K_iiProportional coefficient and integral coefficient of PI regulator for controlling system zero sequence current;

step S11, obtaining the expected value of the system six-phase voltage according to the formula (31)

Wherein the content of the first and second substances,a six-phase voltage desired value;

step S12 is to obtain six-phase voltage u corresponding to all 64 switching states that the six-phase inverter can output, from the formula (32)_AO～u_FO：

Wherein, U_DCIs a DC bus voltage, S_i1(i is a to f) represents that the upper tube of the ith phase bridge arm of the inverter is conducted and the lower tube is turned off; otherwise, S_iWhen the upper tube of the ith phase bridge arm of the inverter is turned off, the lower tube of the ith phase bridge arm of the inverter is turned on, the upper tube of the ith phase bridge arm of the inverter is turned off;

step S13, obtaining cost function values corresponding to 64 switch states according to the formula (33), and taking the switch state with the minimum cost function value to act on the system in the next control period;

Technical Field

The invention relates to a control method for predicting torque of a dual-motor driving system without a weight coefficient.

Background

A six-phase series three-phase dual Permanent Magnet Synchronous Motor (PMSM) driving system is a commonly used dual-Motor driving system. The six-phase series three-phase double-permanent-magnet synchronous motor driving system respectively connects U, V, W phases of a three-phase PMSM with AD, BE and CF of the six-phase PMSM, and realizes independent decoupling control of two PMSMs by a single inverter by utilizing the redundant freedom degree of the six-phase PMSM. Compared with the traditional single PMSM controlled by a single inverter, the six-phase series three-phase double PMSM driving system has the advantages of small driving system volume, low cost, easiness in realizing feedback braking and the like, and has wide application prospects in the industries of steel smelting, rewinders, electric automobiles, textile manufacturing and the like.

And (2) based on a motor system prediction model, performing traversal calculation on Torque and stator flux linkage amplitude values of the motor in the next Control period after different voltage vectors in the applied alternative voltage vector set act on the system, thereby obtaining different cost function values, and enabling the voltage vector with the minimum cost function to act on the next Control period of the system as an optimal voltage vector. The predicted torque control has the advantages of simple structure, intuitive realization, quick dynamic response, easy inclusion of constraint conditions and the like, and is a novel advanced control technology.

Compared with the traditional direct torque control method based on a switching vector table, the six-phase series three-phase dual PMSM driving system adopting the predicted torque control has more excellent steady-state characteristics and dynamic characteristics. However, the traditional cost function based on the torque and stator flux linkage amplitude errors needs to traverse and calculate the torque and stator flux linkage amplitude of the next control period of the two PMSM after different voltage vectors in the applied alternative voltage vector set act on the system, and the calculation amount of the predicted torque control algorithm is greatly increased while the control algorithm is complex. Meanwhile, because the torque and stator flux linkage amplitude dimensions and magnitudes are different, a weight coefficient needs to be introduced into the cost function, the setting work of the weight coefficient is complicated and lacks design basis, and the application of the predictive torque control is limited to a certain extent.

In a six-phase series three-phase dual PMSM driving system, the system has 5 degrees of freedom, and 4 degrees of freedom are needed for controlling the torque and the stator flux linkage amplitude of two PMSMs. In the system, if the last 1 degree of freedom is not effectively controlled, the zero sequence current of the system is overlarge. The large zero sequence current can cause the problems of phase current distortion, system loss increase and the like.

Aiming at the problems, the invention provides a predictive torque control method without a weight coefficient for a six-phase series three-phase double PMSM driving system, so that the accurate control of two PMSM torques, stator flux linkage amplitude and effective suppression of zero-sequence current are realized, the weight coefficient in a cost function is eliminated, and the calculated amount of a predictive torque control algorithm is greatly reduced.

Disclosure of Invention

The invention aims to provide a method for controlling the predicted torque of a dual-motor driving system without a weight coefficient, which realizes the accurate control of the torque and the stator flux linkage amplitude of two PMSMs and the effective inhibition of zero-sequence current; eliminating the weight coefficient of the cost function in the predictive torque control algorithm; and thirdly, reducing the calculation amount of the predicted torque control algorithm.

In order to achieve the purpose, the technical scheme of the invention is as follows: a double-motor driving system prediction torque control method without weight coefficients provides a six-phase series three-phase double-PMSM driving system, obtains expected voltage and zero sequence voltage expected values of two PMSMs under a static coordinate system through a PI regulator according to information including six-phase current, direct current bus voltage and position angle obtained by system sampling, and further obtains expected voltage and zero sequence voltage expected values through T₆The inverse transformation of the matrix obtains the expected value of the six-phase voltage, and the 64 switch states are substituted into the cost function based on the six-phase voltage error, so that the cost function value is the mostThe small switch state acts on the next cycle of the system.

In an embodiment of the present invention, the method of the present invention is specifically implemented as follows:

step S1, using constant power transformation matrix T₆The six-phase current i obtained by sampling_A～i_FConverted into current i on α 1 β 1, α 2 β 2 and o1o2 coordinate systems_α1、i_β1、i_α2、i_β2、i_o1、i_o2：

Wherein i_α1、i_β1、i_α2、i_β2、i_o1、i_o2Current on α 1, β 1, α 2, β 2, o1 and o2 axes respectively, i_o1、i_o2For two zero sequence currents, i is because the neutral point of the three-phase PMSM is not led out_o1Is always 0;

step S2, obtaining stator flux psi of the two PMSMs on the static coordinate system according to the stator flux current model or the stator flux voltage model_sα1、ψ_sβ1、ψ_sα2、ψ_sβ2；ψ_sα1、ψ_sβ1、ψ_sα2、ψ_sβ2Stator flux linkages on α 1, β 1, α 2, β 2 axes, respectively;

1) if a current model of the stator flux linkage is used, the stator flux linkage psi_sα1ψ_sβ1、ψ_sα2ψ_sβ2Comprises the following steps:

wherein psi_f1、ψ_f2Permanent magnet flux linkage for two PMSM; l is_sσ1Is the self leakage inductance, L, of a six-phase PMSM phase winding_sm1＝(L_dm1+L_qm1)/2，L_rs1＝(L_dm1-L_qm1)/2，L_dm1、L_qm1The permanent magnet synchronous motor is characterized by comprising six-phase PMSM phase windings, a main magnetic flux direct shaft inductor and a quadrature axis inductor respectively; l is_sσ2Is the self leakage inductance, L, of the three-phase PMSM phase winding_sm2＝(L_dm2+L_qm2)/2，L_rs2＝(L_dm2-L_qm2)/2，L_dm2、L_qm2For main flux direct-axis and quadrature-axis inductances, theta, of three-phase PMSM phase windings_r1Is the angle between the d1 axis and the α 1 axis, theta_r2Is the angle between the d2 axis and the α 2 axis;

2) if a voltage model of the stator flux linkage is used, the stator flux linkage psi_sα1ψ_sβ1、ψ_sα2ψ_sβ2Comprises the following steps:

step S3, obtaining two PMSM torques T according to the formula (6) and the formula (7)_e1、T_e2：

T_e1＝p₁(ψ_sα1i_β1-ψ_sβ1i_α1)(6)

T_e2＝p₂(ψ_sα2i_β2-ψ_sβ2i_α2)(7)

Wherein p is₁Is six-phase PMSM pole pair number, p₂The number of pole pairs of the three-phase PMSM is shown; t is_e1For six-phase PMSM torque, T_e2Three-phase PMSM torque;

step S4, obtaining two PMSM stator flux linkage amplitude psi according to the formula (8) and the formula (9)_s1、ψ_s2：

Wherein psi_s1The six-phase PMSM stator flux linkage amplitude value; psi_s2The amplitude of the flux linkage of the three-phase PMSM stator is obtained;

step S5, according to the formula (10) and the formula (11), the given torque T of the two PMSMs is obtained by the PI regulators controlling the rotating speeds of the two PMSMs^* _e1、T^* _e2(ii) a Given value psi of six-phase PMSM stator flux linkage amplitude^* _s1Is composed ofGiven value psi of three-phase PMSM stator flux linkage amplitude^* _s2Is composed of

Wherein, ω is_r1、ω_r1A given electrical angular velocity and an actual electrical angular velocity for the six-phase PMSM; omega_r2、ω_r2A given electrical angular velocity and an actual electrical angular velocity for the six-phase PMSM; k_vp1、K_vi1Proportional coefficient and integral coefficient for controlling PI regulator of six-phase PMSM; k_vp2、K_vi2Proportional coefficients and integral coefficients for controlling a PI regulator of the three-phase PMSM;

step S6, obtaining the torque and stator flux linkage amplitude error delta T of the two PMSMs according to the formulas (12) to (15)_e1、Δψ_s1、ΔT_e2、Δψ_s2：

Step S7, adopting fixed parameter PI controller or variable ratio coefficient PI controller to calculate the torque angle variation delta needed for accurately eliminating the two PMSM torque and stator flux linkage amplitude errors₁、Δ₂；

1) If a fixed parameter PI controller is used, the amount of change in torque angle Δ is based on equations (16) and (17)₁、Δ₂Comprises the following steps:

Δ₁＝K_p1ΔT_e1+K_i1∫ΔT_e1dt(16)

Δ₂＝K_p2ΔT_e2+K_i2∫ΔT_e2dt(17)

wherein, K_p1、K_i1Proportional coefficient and integral coefficient of PI regulator for controlling six-phase PMSM torque; k_p2、K_i2Proportional coefficient and integral coefficient of PI regulator for controlling three-phase PMSM torque;

2) if a variable scale coefficient PI controller is adopted, the calculation model is as follows:

2.1) obtaining the scaling factor coefficients of the two PMSMs according to the formula (18) to the formula (21)

Wherein the content of the first and second substances,₁is a six-phase PMSM torque angle,₂is a three-phase PMSM torque angle; l is_d1＝L_sσ1+3L_sm1+3L_rs1Is a six-phase PMSM planar d-axis inductor, L_q1＝L_sσ1+3L_sm1-3L_rs1A six-phase PMSM plane q-axis inductor; l is_d2＝L_sσ1+2L_sσ2+3L_sm2+3L_rs2Is a three-phase PMSM planar d-axis inductor, L_q2＝L_sσ1+2L_sσ2+3L_sm2-3L_rs2A three-phase PMSM plane q-axis inductor;

2.2) obtaining the torque angle variation delta required for accurately eliminating the two PMSM torques and the stator flux linkage amplitude error according to the formula (22) and the formula (23)₁、Δ₂：

2.3) simultaneously, an integral link of the torque error adopts integral separation, and the integral separation only acts when the torque error is small and is used for compensating the inaccuracy of the parameters of the two PMSM;

step S8, obtaining expected values of stator flux linkage variation in α 1 β 1 and α 2 β 2 coordinate systems according to the formula (24) to the formula (27)

Wherein the content of the first and second substances,the expected value of the variation of the stator flux linkage under the α 1 β 1 coordinate system is obtained;is the expected value of the variation of the stator flux linkage in the α 2 β 2 coordinate system, T_sIs a control period; omega_r1、ω_r2Electrical angular velocities of the six-phase PMSM and the three-phase PMSM, respectively;

step S9, obtaining expected values of the stator voltage under the coordinate system α 1 β 1 and α 2 β 2 according to the formula (28) and the formula (29)

Wherein the content of the first and second substances,the expected value of the stator voltage under the coordinate system of α 1 β 1;the expected value of the stator voltage under a α 2 β 2 coordinate system;

step S10, obtaining a zero sequence voltage given value u according to a formula (30)^* _o2：

Wherein, K_ip、K_iiProportional coefficient and integral coefficient of PI regulator for controlling system zero sequence current;

step S11, obtaining the expected value of the system six-phase voltage according to the formula (31)

Wherein the content of the first and second substances,a six-phase voltage desired value;

step S12 is to obtain six-phase voltage u corresponding to all 64 switching states that the six-phase inverter can output, from the formula (32)_AO～u_FO：

Wherein, U_DCIs a DC bus voltage, S_i1(i is a to f) represents that the upper tube of the ith phase bridge arm of the inverter is conducted and the lower tube is turned off; otherwise, S_iWhen the upper tube of the ith phase bridge arm of the inverter is turned off, the lower tube of the ith phase bridge arm of the inverter is turned on, the upper tube of the ith phase bridge arm of the inverter is turned off;

step S13, obtaining cost function values corresponding to 64 switch states according to the formula (33), and taking the switch state with the minimum cost function value to act on the system in the next control period;

compared with the prior art, the invention has the following beneficial effects: 1) the accurate control of the torque and the flux linkage amplitude of the two PMSM and the effective inhibition of the zero-sequence current are realized; 2) the cost function is constructed by directly utilizing the expected value of the six-phase voltage and the error of the six-phase voltage output by the inverter, the weight coefficient is eliminated, and the complicated weight coefficient setting work is omitted; 3) the torque and stator flux linkage amplitude of two PMSMs in the next period after different voltage vectors are acted are not required to be calculated in a traversing mode, the optimal voltage vector is selected by directly utilizing a cost function based on six-phase voltage errors, and the calculation amount of a predictive torque control algorithm is greatly reduced.

Drawings

FIG. 1 is a block diagram of a control strategy according to the present invention.

FIG. 2 is a hardware configuration example of a driving system according to the present invention.

Fig. 3 shows a winding connection mode of a six-phase series three-phase dual PMSM driving system according to the invention.

Fig. 4 is a six-phase PMSM plane.

Fig. 5 is a three-phase PMSM plane.

Fig. 6 is a calculation of flux linkage variation in a six-phase PMSM plane.

Fig. 7 shows the flux linkage variation calculation of the stator in the plane of the three-phase PMSM.

FIG. 8 is a flow chart of the predicted torque control algorithm of the present invention.

Detailed Description

The technical scheme of the invention is specifically explained below with reference to the accompanying drawings.

The invention provides a control method for predicting torque of a dual-motor driving system without a weight coefficient. The purpose has three aspects: firstly, the accurate control of the torque and the stator flux linkage amplitude of two PMSM and the effective inhibition of zero sequence current are realized; eliminating the weight coefficient of the cost function in the predictive torque control algorithm; and thirdly, reducing the calculation amount of the predicted torque control algorithm. According to the six-phase current, the DC bus voltage, the position angle and other information obtained by system samplingThe PI passing regulator obtains expected voltage and zero sequence voltage expected values of two PMSMs under a static coordinate system, and then the expected voltage and the zero sequence voltage are subjected to T₆The inverse transformation of the matrix results in the desired values for the six-phase voltages. And substituting 64 switching states into a cost function based on the six-phase voltage error, and enabling the switching state with the minimum cost function value to act on the next system period. The specific explanation is as follows. The structure block diagram of the system and the control strategy provided by the invention is shown in figure 1. Using T in equation 2₆Matrix, six-phase current i obtained by sampling_A～i_FConverted into current i on α 1 β 1, α 2 β 2 and o1o2 coordinate systems_α1i_β1、i_α2i_β2、 i_o1i_o2(ii) a Obtaining the stator flux linkage psi of the two PMSMs on the static coordinate system according to the stator flux linkage current models of the formula 3 and the formula 4 or the stator flux linkage voltage models of the formula 6 and the formula 7_sα1ψ_sβ1、ψ_sα2ψ_sβ2. Obtaining two PMSM torques T according to the formula 9 and the formula 10_e1、T_e2(ii) a Obtaining two PMSM stator flux linkage amplitude psi according to formula 11 and formula 12_s1、ψ_s2(ii) a According to the formula 41 and the formula 42, given torques T of the two PMSMs are obtained by a PI regulator for controlling the rotating speeds of the two PMSMs^* _e1、T^* _e2(ii) a Given value psi of six-phase PMSM stator flux linkage amplitude^* _s1Is composed ofGiven value psi of three-phase PMSM stator flux linkage amplitude^* _s2Is composed ofObtaining the torque and stator flux linkage amplitude error delta T of the two PMSMs according to the formulas 45 to 48_e1、Δψ_s1、ΔT_e2、Δψ_s2(ii) a According to 49 and formula 50, a PI controller with fixed parameters is adopted to obtain the torque angle variation delta required for calculating and accurately eliminating the two PMSM torque and stator flux linkage amplitude errors₁、Δ₂Or a better proportional coefficient PI controller can be adopted to obtain delta₁、Δ₂: obtaining the scaling factor of the two PMSMs according to the formula 37 to the formula 40Obtaining Δ from equations 43 and 44₁、Δ₂Meanwhile, the integral link of the torque error adopts integral separation, only acts when the torque error is small, is used for compensating the inaccuracy of the parameters of the two PMSM, and obtains the expected values of the stator flux linkage variable quantity under α 1 β 1 and α 2 β 2 coordinate systems according to a formula 51-a formula 54Obtaining α 1 β 1, α 2 β 2 expected values of stator voltage under coordinate system according to formula 55 and formula 56Obtaining the given value u of the zero sequence voltage according to the formula 57^* _o2(ii) a The expected value of the system six-phase voltage is obtained according to equation 58Obtaining six-phase voltage u corresponding to 64 switching states all of which can be output by the six-phase inverter according to formula 61_AO～u_FO(ii) a And obtaining cost function values corresponding to 64 switch states according to a formula 62, and taking the switch state with the minimum cost function value to act on the system in the next control period.

An example of a hardware configuration of a drive system embodying the present invention is shown in fig. 2. The method comprises the following steps: the device comprises a voltage regulator, a three-phase uncontrollable rectifying circuit, a large filtering capacitor, a direct current bus voltage detection circuit, a six-phase inverter, an isolation driving circuit, a six-phase winding current detection circuit, a six-phase PMSM, a three-phase PMSM, an encoder, a DSP, a CPLD, a human-computer interaction interface, a fault protection circuit, an AD conditioning circuit and the like. The power tube in the inverter adopts IGBT or MOSFET. The six-phase winding current detection circuit consists of a Hall current sensor and an operational amplifier circuit, and an output signal is input into the DSP through the AD conditioning circuit. The direct current bus voltage detection circuit consists of a Hall voltage sensor and an operational amplifier circuit, and an output signal also needs to be input into the DSP through the AD conditioning circuit. The detection signals of the six-phase winding current and the direct-current bus voltage also need to pass through the fault protection circuit, when the direct-current bus voltage and the winding current are abnormal, the fault protection circuit outputs signals to the CPLD, and the CPLD blocks the PWM output signals to turn off all the switch tubes. The rotor position angles of the two PMSM's are detected by two incremental photoelectric encoders. The six-phase winding current detection signal, the direct current bus voltage detection signal and the rotor position angle signals of the two PMSMs are input into the DSP, the DSP outputs a control signal of the switching tube according to the detected signal and the control strategy of the invention, then the CPLD detects whether the control signal can cause the switching tube to generate direct connection danger or not, and if not, the CPLD outputs the control signal to the isolation driving circuit to control the power switching tube in the inverter to act so as to control the two PMSMs.

The connection mode of the six-phase series three-phase dual-PMSM driving system windings is shown in figure 3, wherein the A-F phases are six-phase PMSM phase windings, the U-W phases are three-phase PMSM phase windings, and the windings of the two PMSM phases are symmetrically distributed in space. The three-phase PMSM adopts a star connection mode. U, V, W phases of the three-phase PMSM are connected to AD, BE, and CF of the six-phase PMSM, respectively. The power current components of the six-phase PMSM are mutually offset due to opposite phases at the connection part of the six-phase PMSM and the three-phase PMSM winding, so that the three-phase PMSM is not influenced; the power current component of the three-phase PMSM equally flows through the two-phase windings with opposite phases of the six-phase PMSM, and has no influence on the six-phase PMSM, so that the decoupling control of the two PMSMs is realized.

FIGS. 4 and 5 are plane definitions of electromechanical energy conversion for six-phase PMSM and three-phase PMSM in FIG. 4, α 1 β 1 and d1q1 coordinate systems are stationary coordinate system and synchronous rotating coordinate system of six-phase PMSM plane, A-F are winding axis of each phase of six-phase PMSM, theta_r1Is the angle between the d1 axis and the α 1 axis,₁is the torque angle, ω_r1、u_s1、i_s1、ψ_s1、ψ_f1The electrical angular velocity, the stator voltage vector, the stator current vector, the stator flux linkage vector and the rotor flux linkage vector of the six-phase PMSM are respectively. The three-phase PMSM plane variable definition in fig. 5 is similar to that of fig. 4. In addition, the system has 1 zero sequence which does not participate in electromechanical energy conversion, and the zero sequence is called a zero sequence plane.

According to the connection mode of the two PMSM windings in fig. 3, the expression of the output voltage of the six-phase inverter can be obtained as follows:

wherein u is_AO～u_FOThe voltages from the A-F output ends of the six-phase inverter, namely the A-F phase winding input ends of the six-phase PMSM, to the neutral point O of the three-phase PMSM are respectively referred to as A-F phase voltages; i.e. i_A～i_FPhase current for a six-phase PMSM; r_s1Resistance, R, of each phase winding of a six-phase PMSM_s2The resistance of each phase winding of the three-phase PMSM; psi_sA～ψ_sFEach phase of stator flux linkage is six-phase PMSM; psi_sU～ψ_sWThe magnetic flux linkage is a stator magnetic flux linkage of each phase of the three-phase PMSM.

Constant power transformation matrix T using equation 2₆And transforming the mathematical model of the dual PMSM drive system from an ABCDEF natural coordinate system to an α 1 β 1 α 2 β 2o1o2 stationary coordinate system.

The flux linkage model under the coordinate systems of alpha 1 beta 1, alpha 2 beta 2 and o1o2 is as follows:

wherein psi_sα1、ψ_sβ1、ψ_sα2、ψ_sβ2、ψ_so1、ψ_so2Is stator flux linkage on α 1, β 1, α 2, β 2, o1 and o2 axes i_α1、i_β1、i_α2、i_β2、i_o1、i_o2Current on α 1, β 1, α 2, β 2, o1 and o2 axes, #_f1Permanent magnet flux linkage, psi, for a six-phase PMSM_f2A permanent magnet flux linkage of a three-phase PMSM; l is_sσ1Is the self leakage inductance, L, of a six-phase PMSM phase winding_sm1＝(L_dm1+L_qm1)/2， L_rs1＝(L_dm1-L_qm1)/2，L_dm1、L_qm1The permanent magnet synchronous motor is characterized by comprising six-phase PMSM phase windings, a main magnetic flux direct shaft inductor and a quadrature axis inductor respectively; l is_sσ2Is the self leakage inductance, L, of the three-phase PMSM phase winding_sm2＝(L_dm2+L_qm2)/2，L_rs2＝(L_dm2-L_qm2)/2，L_dm2、L_qm2The three-phase PMSM phase winding is provided with a main magnetic flux direct-axis inductor and a quadrature-axis inductor.

The voltage model under the coordinate systems of alpha 1 beta 1, alpha 2 beta 2 and o1o2 is as follows:

wherein u is_α1、u_β1、u_α2、u_β2、u_o1、u_o2Voltages on the α 1, β 1, α 2, β 2, o1, o2 axes.

The torques of the two PMSM are:

T_e1＝p₁(ψ_sα1i_β1-ψ_sβ1i_α1) (formula 9)

T_e2＝p₂(ψ_sα2i_β2-ψ_sβ2i_α2) (formula 10)

Wherein, T_e1For six-phase PMSM torque, T_e2Three-phase PMSM torque; p is a radical of₁Is six-phase PMSM pole pair number, p₂Is the three-phase PMSM pole pair number.

The stator flux linkage amplitudes of the two PMSMs are respectively as follows:

wherein psi_s1The six-phase PMSM stator flux linkage amplitude value; psi_s2The three-phase PMSM stator flux linkage amplitude.

The quantities in the α 1 β 1 coordinate system are transformed to the d1q1 coordinate system using the six-phase PMSM plane rotation transformation matrix R (θ R1) of equation 13, and the quantities in the α 2 β 2 coordinate system are transformed to the d2q2 coordinate system using the three-phase PMSM plane rotation transformation matrix R (θ R2) of equation 14.

The flux linkage model under the d1q1 and d2q2 coordinate systems is as follows:

wherein psi_sd1、ψ_sq1、ψ_sd2、ψ_sq2Stator flux linkages on d1, q1, d2 and q2 axes; i.e. i_d1、i_q1、i_d2、i_q2Currents on d1, q1, d2 and q2 axes; l is_d1＝L_sσ1+3L_sm1+3L_rs1Is a six-phase PMSM planar d-axis inductor, L_q1＝L_sσ1+3L_sm1-3L_rs1A six-phase PMSM plane q-axis inductor; l is_d2＝L_sσ1+2L_sσ2+3L_sm2+3L_rs2Is a three-phase PMSM planar d-axis inductor, L_q2＝L_sσ1+2L_sσ2+3L_sm2-3L_rs2Is a three-phase PMSM plane q-axis inductor.

The voltage model under the d1q1 and d2q2 coordinate systems is as follows:

wherein u is_d1、u_q1、u_d2、u_q2Voltages on d1, q1, d2 and q2 axes.

The torques of the two PMSM are:

T_e1＝p₁(ψ_sd1i_q1-ψ_sq1i_d1) (formula 19)

T_e2＝p₂(ψ_sd2i_q2-ψ_sq2i_d2) (formula 20)

The stator flux linkage amplitudes of the two PMSMs are respectively as follows:

as can be seen from fig. 4 and 5, the stator flux linkage in the d1q1 and d2q2 coordinate systems can be further represented as:

ψ_sd1＝ψ_s1cos₁(formula 23)

ψ_sq1＝ψ_s1sin₁(formula 24)

ψ_sd2＝ψ_s2cos₂(equation 25)

ψ_sq2＝ψ_s2sin₂(formula 26)

Wherein the content of the first and second substances,

wherein, theta_s1The angle of the six-phase PMSM stator flux linkage vector is shown; theta_s2Is the angle of the flux linkage vector of the three-phase PMSM stator.

Substituting equations 23 and 24 into 15 and equations 25 and 26 into 16 yields:

substituting equations 29 and 30 into equation 19 and equations 31 and 32 into equation 20, the torques of the two PMSM are:

the differential with respect to time is obtained on both sides of equation 33 and equation 34:

wherein the content of the first and second substances,

according to the formulas 37 to 40, the scaling coefficients of the two PMSMs are shownRelated to the operating state of the two PMSM's. When the amplitude of the stator flux linkage is a given value,the torque loop can adopt a variable proportion type PI regulator in order to accelerate the torque response of the system in the dynamic process.

According to the formula 41 and the formula 42, given torques T of the two PMSMs are obtained by a PI regulator for controlling the rotating speeds of the two PMSMs^* _e1、T^* _e2(ii) a Given value psi of six-phase PMSM stator flux linkage amplitude^* _s1Is composed ofGiven value psi of three-phase PMSM stator flux linkage amplitude^* _s2Is composed of

Wherein, ω is_r1、ω_r1A given electrical angular velocity and an actual electrical angular velocity for the six-phase PMSM; omega_r2、ω_r2A given electrical angular velocity and an actual electrical angular velocity of the six-phase PMSM. K_vp1、K_vi1Proportional coefficient and integral coefficient for controlling PI regulator of six-phase PMSM; k_vp2、K_vi2To control the proportionality and integral coefficients of a PI regulator of a three-phase PMSM.

Obtaining the torque angle variation quantity required when the two PMSM torques and the stator flux linkage amplitude errors can be accurately eliminated according to the formula 35 and the formula 36:

wherein the content of the first and second substances,

wherein, Delta T_e1、Δψ_s1For six-phase PMSMMoment error and stator flux linkage amplitude error; delta T_e2、Δψ_s2The torque error and the stator flux linkage amplitude error of the three-phase PMSM are shown.

Meanwhile, an integral link of the torque error adopts integral separation, and the integral separation only acts when the torque error is small and is used for compensating the inaccuracy of the parameters of the two PMSM.

In addition, Δ is simplified if necessary₁、Δ₂The calculation process of (2) can also adopt a fixed parameter PI controller, the given value is a given torque value, the feedback value is an actual torque value, namely:

Δ₁＝K_p1ΔT_e1+K_i1∫ΔT_e1dt (equation 49)

Δ₂＝K_p2ΔT_e2+K_i2∫ΔT_e2dt (equation 50)

Wherein, K_p1、K_i1Proportional coefficient and integral coefficient of PI regulator for controlling six-phase PMSM torque; k_p2、K_i2Proportional coefficient and integral coefficient of PI regulator for controlling six-phase PMSM torque;

schematic diagrams for calculating flux linkage variation of the two PMSM in-plane stators are shown in FIGS. 6 and 7. The expected values of the stator flux linkage variation under the coordinate systems of alpha 1 beta 1 and alpha 2 beta 2 are as follows:

wherein the content of the first and second substances,the expected value of the variation of the stator flux linkage under the α 1 β 1 coordinate system is obtained;is the expected value of the variation of the stator flux linkage in the α 2 β 2 coordinate system, T_sIs a control cycle.

In order to make the stator flux linkage variation amount track the expected value in the α 1 β 1 and α 2 β 2 coordinate systems, the expected value of the stator voltage in the α 1 β 1 and α 2 β 2 coordinate systems is obtained according to equations 6 and 7 as follows:

wherein the content of the first and second substances,the expected value of the stator voltage under the coordinate system of α 1 β 1;the expected value of the stator voltage under the coordinate system of α 2 β 2.

The leakage inductance of the six-phase PMSM is very small, very small zero-sequence voltage can generate very small zero-sequence current, the existence of the zero-sequence current can not only influence the THD of the six-phase current, but also cause the system loss to be increased, and the overall efficiency of the double-PMSM driving system is reduced. Introducing a zero-sequence current PI regulator, setting value as 0, feedback value as zero-sequence current value, and outputting as zero-sequence voltage setting value u^* _o2. When the zero sequence voltage output by the inverter is the value, the zero sequence current is controlled to be 0.

Wherein, K_ip、K_iiProportional coefficient and integral coefficient of PI regulator for controlling system zero sequence current.

In summary, the expected values of the six-phase voltages of the system are:

wherein the content of the first and second substances,the six-phase voltage desired value.

As can be taken from fig. 3, the six-phase voltages can also be represented as:

wherein, U_DCIs a DC bus voltage u_NORepresenting the voltage from DC bus voltage ground N to neutral O, S_i1(i is a to f) represents that the upper tube of the ith phase bridge arm of the inverter is conducted and the lower tube is turned off; otherwise, S_iAnd 0 represents that the upper tube of the ith phase bridge arm of the inverter is turned off and the lower tube of the ith phase bridge arm of the inverter is turned on.

Since the neutral point of the system is not led out, u is_AO+u_BO+u_CO+u_DO+u_EO+u_FOWhen 0 is obtained, formula 59 is substituted to obtain:

substituting equation 60 into equation 59 yields:

according to a traditional cost function based on torque and stator flux linkage amplitude errors, two PMSM torque and stator flux linkage amplitude in the next period after different voltage vectors in an alternative voltage vector set are applied need to be calculated in a traversing mode according to prediction models of the two PMSM, the calculation amount is huge, and meanwhile, the weight coefficient setting work inevitably existing in the cost function is troublesome. Aiming at the problems, the cost function based on the six-phase voltage error is established according to the formula 62, the weight coefficient in the traditional cost function is eliminated by the cost function, and the complicated weight coefficient setting work is omitted; meanwhile, the cost does not need to predict the torque and the stator flux linkage amplitude of the next period of the two PMSMs, and the calculated amount of a predicted torque control algorithm is greatly reduced.

The six-phase inverter can output 64 different switching states, the different switching states correspond to different six-phase voltage values, and six-phase voltage u corresponding to all the switching states which can be output by the six-phase inverter can be obtained according to a formula 61_AO～u_FO. And substituting the 64 groups of six-phase voltage values into a cost function for calculation, wherein the switching state with the minimum cost function value is the optimal switching state, and the optimal switching state acts on the next period of the system.

A flow chart of the proposed predicted torque control algorithm is shown in fig. 8.

The above are preferred embodiments of the present invention, and all changes made according to the technical scheme of the present invention that produce functional effects do not exceed the scope of the technical scheme of the present invention belong to the protection scope of the present invention.