Sensorless global delay compensation strategy for permanent magnet synchronous motor under square wave field weakening

Abstract

The present invention discloses a sensorless global delay compensation strategy for a permanent magnet synchronous motor under square wave field weakening. Specifically, the strategy comprises the following steps: using a sliding mode observer as a position observation model to obtain an actual rotor position estimate, forming a speed loop; performing coordinate transformation based on the ideal and actual estimated rotor position angles to calculate the corresponding relationship between the ideal and lag signals of the quadrature and direct axis feedback currents; obtaining the ideal and actual quadrature and direct axis feedback current signals, performing error correction, and performing online PI adjustment on the estimated error of the obtained phase lag angle; and obtaining a final voltage output command based on voltage vector angle control of a single q-axis current regulator to achieve global delay compensation. Based on a position sensorless closed-loop structure with single current regulation, the method of the present invention performs real-time compensation of the rotor position during coordinate transformation, improving the performance of sensorless field weakening control under square wave modulation and further enhancing position estimation accuracy.

Description

Technical Field

[0001] The present invention belongs to the technical field of AC motor transmission control, and in particular relates to a sensorless global delay compensation strategy for a permanent magnet synchronous motor under square wave magnetic weakening. Background Art

[0002] EMU traction drive units feature high voltage, high current, and a wide operating speed range. Furthermore, they require the traction motors to operate under low switching frequencies and square wave conditions. With the rapid development of high-speed rail trains, permanent magnet synchronous motors (PMSMs) with a wide speed range, high power density, and low energy consumption have become a research hotspot in the EMU traction field. However, existing EMU traction systems all use mechanical position / speed sensors to obtain motor position or speed signals. During high-speed train operation, the complex electromagnetic environment and intense vibrations can easily cause mechanical sensor failure, leading to traction system failures and high torque shocks. In severe cases, this can damage critical components such as bearings, gears, and motors, compromising train safety. Sensorless drive technology can fundamentally eliminate this safety hazard, offering advantages such as strong anti-interference capabilities, high integration, and a long service life.

[0003] In permanent magnet traction drive systems, system delays in vector closed-loop control primarily include inverter nonlinearity, sampling and filtering delays, and algorithm execution time. The latter two can be collectively referred to as computational delay. To achieve good inverter voltage output performance at low switching frequencies (<=500Hz) and fully utilize the bus voltage, single-pulse square wave modulation is typically used when the motor enters the field weakening region—that is, when the speed exceeds the rated speed and the terminal voltage amplitude is saturated. In this case, the impact of inverter nonlinearity delays, such as dead time, within a single fundamental voltage cycle is negligible. As the motor speed increases in the field weakening region, the magnetic field orientation error caused by computational delays can lead to rotor position estimation bias, degrading sensorless control performance. In particular, when the inverter operates in the square wave modulation region, rotor position estimation is performed using a motor model. The computational delay of a single pulse at low carrier-to-carrier ratios can cause significant amplitude attenuation and phase lag in the fundamental signal. In this case, the dynamics and stability of the position observer are affected by the dynamic coupling of the current regulator. Therefore, in the permanent magnet traction system, combining the characteristics of single q-axis current regulation weak magnetic control and square wave modulation to compensate for the calculation delay has important practical significance for improving the position sensorless control performance. Summary of the Invention

[0004] The purpose of the present invention is to provide a sensorless global delay compensation strategy for a permanent magnet synchronous motor under square wave magnetic weakening. The current lag phase angle is estimated by the q-axis current error, thereby compensating the estimated rotor position angle. Combined with the error correction of the current amplitude, the performance of the sensorless magnetic weakening control under square wave modulation is improved.

[0005] The technical solution adopted by the present invention is a sensorless global delay compensation strategy for a permanent magnet synchronous motor under square wave field weakening, which is specifically implemented according to the following steps:

[0006] Step 1: Use the sliding mode observer in the rotating coordinate system as the position observation model to obtain the actual rotor position estimate Thus forming a speed loop;

[0007] Step 2: Perform coordinate transformation based on the ideal and actual estimated rotor position angles to calculate the corresponding relationship between the ideal signal and the hysteresis signal of the quadrature and direct axis feedback currents;

[0008] Step 3: Obtain the ideal and actual signals of the quadrature and direct axis feedback currents through a single q-axis current regulation position sensorless closed-loop structure;

[0009] Step 4, performing amplitude error correction on the feedback current to calculate the estimated error of the phase lag angle;

[0010] Step 5: Design an adaptive PI compensator to perform online PI adjustment on the estimated error of the phase lag angle, thereby eliminating the phase lag error;

[0011] Step 6: Based on the voltage vector angle control of the single q-axis current regulator, the final voltage output command is obtained to achieve global delay compensation.

[0012] The present invention is also characterized in that:

[0013] In step 1, specifically:

[0014] Step 1.1, establish the dq sliding mode observer model, as shown in formula (1);

[0015]

[0016] in, and are the stator d-axis and q-axis current observation values, ωre is the rotor angular velocity, D is the differential operator, Rs is the stator resistance, and L d is the direct axis component of the stator inductance, L q is the quadrature-axis component of the stator inductance, u d is the direct axis component of the stator voltage, u q is the quadrature-axis component of the stator voltage; Vd and Vq are the control inputs of the sliding mode observer respectively; i d is the direct axis component of the stator current, i q is the quadrature-axis component of the stator current.

[0017] Step 1.2: Low-pass filter the PI module to obtain the estimated rotor angular velocity. As shown in formula (2);

[0018]

[0019] Among them, Kpo is the proportional gain and Kio is the integral gain;

[0020] Estimating the angular velocity of the rotor By performing integration operation, the actual rotor position estimate can be obtained. This forms a speed loop.

[0021] In step 2, specifically:

[0022] In step 2.1, without considering the calculation delay, the ideal stator three-phase current is shown in equation (3);

[0023]

[0024] Among them, i A is the phase A current, i B is the B phase current, i C is the C phase current, I m is the modulus of the current vector, θ ph is the ideal current vector angle;

[0025] At this time, in position sensorless control, the change matrix R from the ABC phase to the dq coordinate system is shown in formula (4);

[0026]

[0027] in, is the ideal estimated rotor position angle;

[0028] According to equation (4), coordinate transformation is performed on equation (3) to obtain the ideal quadrature and direct axis feedback current, as shown in equation (5);

[0029]

[0030] Where, σ is the angle between the current vector I and the d-axis;

[0031] Step 2.2, after considering the calculation delay, the current vector angle and the estimated rotor position angle have a The hysteresis angle is converted from equation (3) to equation (6), and from equation (4) to equation (7);

[0032]

[0033] Among them, I m_delay To consider the current vector modulus after delay, is the current phase lag angle;

[0034]

[0035] in, is the estimated current phase lag angle;

[0036] According to equation (7), equation (6) is transformed into coordinates to obtain the actual quadrature and direct axis feedback current as shown in equation (8);

[0037]

[0038] Among them, i d_delay is the actual direct-axis feedback current; i q_delay is the actual quadrature-axis feedback current;

[0039] In step 2.3, by combining equations (5) and (8), we can obtain the corresponding relationship between the ideal and actual signals of the quadrature and direct axis currents as shown in equation (9);

[0040]

[0041] Among them, M err is the amplitude error caused by the delay, which is written as formula (10) in the dq coordinate system;

[0042]

[0043] In step 3, specifically:

[0044] Step 3.1, for the three-phase current i A 、i B 、i C Sampling is performed, Clark transformation is performed, and the rotor position estimate obtained in step 1 is obtained. Perform Park transformation to obtain the actual quadrature and direct axis feedback current idq_delay;

[0045] Step 3.2, obtain the ideal quadrature and direct axis feedback current signals under the field weakening constraint through the current trajectory planning link of the single q-axis current regulator field weakening control;

[0046] Step 3.3, substitute the dq axis current command under the weak magnetic constraint into equations (9) and (10) to obtain equations (15) and (16);

[0047]

[0048]

[0049] In step 3.2, specifically:

[0050] Step 3.2.1, calculate the preliminary d-axis current command through the MTPA control relationship As shown in formula (11);

[0051]

[0052] Among them, ψ f is the permanent magnet flux of the motor, is the q-axis current command;

[0053] Step 3.2.2: Adjust the current command according to the field weakening control target; specifically:

[0054] Step 3.2.2.1, the d-axis and q-axis feedforward voltages provided by the feedforward link and Calculate the feedback voltage amplitude As shown in formula (12);

[0055]

[0056] Step 3.2.2.2, set the voltage vector magnitude u max The weak magnetic compensation current Δi is obtained by PI adjustment of the error of the feedback voltage amplitude d,wkfd ;

[0057] Step 3.2.2.3, the weak magnetic compensation current Δi output by the voltage closed loop d,wkfd , correct the d-axis current command to the optimal current trajectory, as shown in formula (13);

[0058]

[0059] Step 3.2.2.4, calculate the q-axis current command under the field weakening control constraint through the torque formula As shown in formula (14);

[0060]

[0061] In step 4, specifically:

[0062] Step 4.1, without considering the delay, eliminate the amplitude error of equation (15) to obtain equation (17);

[0063]

[0064] Among them, i d_com is the direct axis current after amplitude error correction, i q_com is the quadrature-axis current after amplitude error correction;

[0065] Step 4.2, in square wave weak magnetic single q-axis current regulation, estimate the value Close to actual value hour, and Then equation (17) can be transformed into equation (18);

[0066]

[0067] Based on formula (18), the estimation error of the phase lag angle can be expressed as formula (19);

[0068]

[0069] In step 5, specifically:

[0070] Step 5.1: Select the estimated error of the phase lag angle and its rate of change as the input of the parameter self-tuning module. The calculation rule in the self-tuning module is shown in formula (20);

[0071]

[0072] in, is the output of the parameter self-tuning module; k 11 、k 21 、k 12 and k 22 is the scale factor, b 11 、b 21 、b 12 and b 22 is the weighting factor;

[0073] Step 5.2, the proportional and integral gains can be calculated using equation (21);

[0074]

[0075] Among them, ξ p is the proportional gain, ξ i is the integral gain;

[0076] Step 5.3, the lag angle is obtained by performing online PI adjustment on the estimated error. When the estimated phase lag angle is compensated to the estimated value of the rotor position, the estimated rotor position angle is given by becomes As shown in formula (22);

[0077]

[0078] in, is the estimated value of the rotor position after compensation.

[0079] In step 6, specifically:

[0080] Step 6.1, the estimated value of the rotor position after compensation obtained in step 5 Re-transform the three-phase current coordinates to obtain the stator current quadrature axis component i q , and for i q Perform amplitude error correction to obtain the q-axis current i after amplitude error correction q_corr ;

[0081] Step 6.2, according to the q-axis current command under the weak magnetic control constraint obtained in step 3.2 The voltage vector angle θ can be obtained by using the q-axis current iq_corr after amplitude error correction obtained in step 6.1: VVA ;

[0082]

[0083] Step 6.3: Calculate the d-axis and q-axis command voltages output under the field-weakening control of the voltage vector angle of the single q-axis current regulator according to equation (24) to achieve global delay compensation for the single q-axis current regulation.

[0084]

[0085] Among them, u max is the voltage vector magnitude, u max =2u dc / π,u dc is the DC bus voltage.

[0086] The beneficial effects of the present invention are:

[0087] 1) Global delay compensation of interior permanent magnet synchronous motor under square wave modulation under field weakening control is realized;

[0088] 2) Design an adaptive PI compensator to further ensure real-time compensation accuracy and system stability;

[0089] 3) The performance of weak magnetic control without position sensor in square wave modulation is improved, and the position estimation accuracy is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] Figure 1 This is a principle block diagram of the sensorless global delay compensation strategy for a permanent magnet synchronous motor under square wave field weakening of the present invention;

[0091] Figure 2 This is a block diagram of the hardware circuit structure of the experimental system used in the sensorless global delay compensation strategy of the permanent magnet synchronous motor under square wave magnetic weakening of the present invention;

[0092] Figure 3 This is a block diagram of the magnetic field weakening control strategy under single q-axis current regulation of the permanent magnet synchronous motor of the present invention;

[0093] Figure 4This is a schematic diagram of the global delay compensation principle based on the q-axis current error in the method of the present invention;

[0094] Figure 5 is a diagram showing the effect of calculation delay on estimated rotor position in the method of the present invention;

[0095] Figure 6 is a structural diagram of the adaptive PI compensator in the method of the present invention;

[0096] Figure 7 It is the PI module in the sensorless global delay compensation strategy of the permanent magnet synchronous motor under square wave magnetic weakening of the present invention;

[0097] Figure 8 This is a waveform diagram of the current performance comparison experiment before and after the delay compensation of the position sensorless single current regulation of the present invention;

[0098] Figure 9 This is a waveform diagram of the rotor position estimation performance comparison experiment before and after position sensorless single current regulation delay compensation of the present invention. DETAILED DESCRIPTION

[0099] The principle block diagram of the sensorless global delay compensation strategy for permanent magnet synchronous motor under square wave magnetic weakening of the present invention is as follows: Figure 1 As shown in Figure 2, this strategy is based on a closed-loop structure without position sensors under single q-axis current regulation. Considering the influence of calculation delay, a delay compensation link based on q-axis current error is designed to improve the rotor position estimation performance, as shown in Figure 2. Figure 4 As shown; specifically follow the steps below:

[0100] Step 1: Use the sliding mode observer in the rotating coordinate system as the position observation model to obtain the actual rotor position estimate Thus forming a speed loop; specifically:

[0101] Step 1.1, establish the dq sliding mode observer model, as shown in formula (1);

[0102]

[0103] in, and are the stator d-axis and q-axis current observation values, ωre is the rotor angular velocity, D is the differential operator, R s is the stator resistance, L d is the direct axis component of the stator inductance, L q is the quadrature-axis component of the stator inductance, u d is the direct axis component of the stator voltage, u q is the quadrature-axis component of the stator voltage; Vd and Vq are the control inputs of the sliding mode observer respectively;

[0104]

[0105] i d is the direct axis component of the stator current, i q is the quadrature axis component of the stator current; the function sgn is is the sliding mode observer gain;

[0106] Step 1.2, through the PI module, such as Figure 7 As shown, the estimated rotor angular velocity can be obtained by low-pass filtering. As shown in formula (2);

[0107]

[0108] Among them, Kpo is the proportional gain and Kio is the integral gain;

[0109] Estimating the angular velocity of the rotor By performing integration operation, the actual rotor position estimate can be obtained. Thus forming a speed loop;

[0110] Step 2: Perform coordinate transformation based on the ideal and actual estimated rotor position angles to calculate the corresponding relationship between the ideal signal and the hysteresis signal of the quadrature and direct axis feedback currents; specifically:

[0111] In step 2.1, without considering the calculation delay, the ideal stator three-phase current is shown in equation (3);

[0112]

[0113] Among them, i A is the ideal phase A current, i B is the ideal B phase current, i C is the ideal C phase current, I m is the modulus of the current vector, θ ph is the ideal current vector angle;

[0114] At this time, in position sensorless control, the change matrix R from ABC to dq coordinate system is shown in formula (4);

[0115]

[0116] in, is the ideal estimated rotor position angle;

[0117] According to equation (4), coordinate transformation is performed on equation (3) to obtain the ideal quadrature and direct axis feedback current, as shown in equation (5);

[0118]

[0119] Where, σ is the angle between the current vector I and the d-axis;

[0120] Step 2.2, after considering the calculation delay, the current vector angle and the estimated rotor position angle have a The hysteresis angle, such as Figure 5 As shown, transform equation (3) into equation (6), and transform equation (4) into equation (7);

[0121]

[0122] Among them, I m_delay To consider the current vector modulus after delay, is the current phase lag angle; i A_delay To consider the A phase current after delay, i B_delay To consider the B phase current after delay, i C_delay To consider the delayed C phase current,

[0123]

[0124] in, is the estimated current phase lag angle;

[0125] According to equation (7), equation (6) is transformed into coordinates to obtain the actual quadrature and direct axis feedback current as shown in equation (8);

[0126]

[0127] Among them, i d_delay is the actual direct-axis feedback current; i q_delay is the actual quadrature-axis feedback current;

[0128] In step 2.3, by combining equations (5) and (8), we can obtain the corresponding relationship between the ideal and actual signals of the quadrature and direct axis currents as shown in equation (9);

[0129]

[0130] Among them, M err is the amplitude error caused by the delay, satisfying M err =I m_delay / I m , which can be written as formula (10) in the dq coordinate system;

[0131]

[0132] Step 3: Obtain the ideal and actual signals of the quadrature and direct axis feedback currents through a single q-axis current regulation position sensorless closed-loop structure; specifically:

[0133] Step 3.1, for the three-phase current i A、i B 、i C Sampling is performed, Clark transformation is performed, and the rotor position estimate obtained in step 1 is obtained. Perform Park transformation to obtain the actual quadrature and direct axis feedback current idq_delay;

[0134] Step 3.2, obtain the ideal signal of the quadrature and direct axis feedback current under the weak magnetic field constraint through the single q-axis current regulator weak magnetic field control current trajectory planning link, such as Figure 3 As shown; specifically:

[0135] Step 3.2.1, calculate the preliminary d-axis current command through the MTPA control relationship As shown in formula (11);

[0136]

[0137] Among them, ψ f is the permanent magnet flux of the motor, is the q-axis current command;

[0138] Step 3.2.2: Adjust the current command according to the field weakening control target; specifically:

[0139] Step 3.2.2.1, the d-axis and q-axis feedforward voltages provided by the feedforward link and Calculate the feedback voltage amplitude As shown in formula (12);

[0140]

[0141] Step 3.2.2.2, set the voltage vector magnitude u max The weak magnetic compensation current Δi is obtained by PI adjustment of the error of the feedback voltage amplitude d,wkfd ;

[0142] Step 3.2.2.3, the weak magnetic compensation current Δi output by the voltage closed loop d,wkfd , correct the d-axis current command to the optimal current trajectory, as shown in formula (13);

[0143]

[0144] Step 3.2.2.4, calculate the q-axis current command under the field weakening control constraint through the torque formula As shown in formula (14);

[0145]

[0146] Step 3.3, substitute the dq axis current command under the weak magnetic constraint into equations (9) and (10) to obtain equations (15) and (16);

[0147]

[0148]

[0149] Step 4: Correct the amplitude error of the feedback current and calculate the estimated error of the phase lag angle; specifically:

[0150] Step 4.1, without considering the delay, eliminate the amplitude error of equation (15) to obtain equation (17);

[0151]

[0152] Among them, i d_com is the direct axis current after amplitude error correction, i q_com is the quadrature-axis current after amplitude error correction;

[0153] Step 4.2: When adjusting the square wave weak magnetic single q-axis current, the q-axis current error has a smaller amplitude fluctuation than the d-axis current error, and is easy to use to estimate the phase lag angle. Close to actual value hour, and Then equation (17) can be transformed into equation (18);

[0154]

[0155] Based on formula (18), the estimation error of the phase lag angle can be expressed as formula (19);

[0156]

[0157] Step 5: Design an adaptive PI compensator to perform online PI adjustment on the estimated phase lag angle to eliminate the phase lag error. Specifically:

[0158] Step 5.1, select the estimated error of the phase lag angle and its rate of change obtained in step 4 as the input of the parameter self-tuning module, such as Figure 6 As shown, the calculation rules in the self-tuning module are shown in formula (20);

[0159]

[0160] in, is the output of the parameter self-tuning module; k 11 、k 21 、k 12 and k22 is the proportional factor, which can macro-adjust the size of the estimation error and its rate of change, b 11 、b 21 、b 12 and b 22 is a weighting factor, which can correct the weighted proportion of the two input quantities from a microscopic perspective. Here, the values of each factor are k 11 =k 12 =5, k 21 =k 22 =1, b 11 =b 21 =b 12 =b 22 =2;

[0161] Step 5.2, the proportional and integral gains can be calculated using equation (21);

[0162]

[0163] Among them, ξ p is the proportional gain, ξ i is the integral gain;

[0164] Step 5.3, the lag angle is obtained by performing online PI adjustment on the estimated error. When the estimated phase lag angle is compensated to the estimated value of the rotor position, the estimated rotor position angle is given by becomes As shown in formula (22);

[0165]

[0166] in, is the estimated value of the rotor position after compensation;

[0167] Step 6: Based on the voltage vector angle control of the single q-axis current regulator, the final voltage output command is obtained to achieve global delay compensation; specifically:

[0168] Step 6.1, the estimated value of the rotor position after compensation obtained in step 5 Re-transform the three-phase current coordinates to obtain the stator current quadrature axis component i q , and for i q Perform amplitude error correction to obtain the q-axis current i after amplitude error correction q_corr ;

[0169] Step 6.2, in the current command tracking link of the single q-axis current regulator weak magnetic control, such as Figure 3 As shown in the figure, the q-axis current command under the weak magnetic control constraint obtained in step 3.2 is The voltage vector angle θ can be obtained by using the q-axis current iq_corr after amplitude error correction obtained in step 6.1: VVA ;

[0170]

[0171] Among them, Kp is the proportional gain and Ki is the integral gain;

[0172] In step 6.3, the d-axis and q-axis command voltages output under the weak magnetic control of the voltage vector angle of the single q-axis current regulator can be calculated according to formula (24), thus realizing the global delay compensation of the single q-axis current regulation.

[0173]

[0174] Among them, u max is the voltage vector magnitude, u max =2u dc / π,u dc is the DC bus voltage;

[0175] When a permanent magnet synchronous motor operates in the high-speed range, it is generally in a square-wave operating condition. The computational delay of a single pulse at a low carrier ratio under square-wave modulation causes the fundamental signal to experience severe amplitude attenuation and phase lag, resulting in large rotor position estimation errors and reduced sensorless field-weakening control performance. To improve the field-weakening control performance of a built-in permanent magnet synchronous motor under sensorless field-weakening control, the present invention proposes a global delay compensation method based on the q-axis current error for sensorless field-weakening control of a built-in permanent magnet synchronous motor under square-wave modulation, thereby improving the sensorless field-weakening control performance under square-wave conditions and enhancing the rotor position estimation performance.

[0176] The system hardware structure of the present invention is as follows Figure 2 As shown, it includes: a rectifier circuit, a filter circuit, a three-phase full-bridge inverter, an IPMSM (internal permanent magnet synchronous motor), an RTLAB real-time controller, an isolation drive circuit, a rotary transformer and a current acquisition circuit; this system uses a rotary transformer to collect real position signals. Figures 8 to 9 For the motor Figure 2 The steady-state performance waveform comparison under square wave modulation when the strategy is not used and after using the strategy under the control of the hardware system shown. Figure 8The following waveforms compare the current performance before and after using this strategy: When this strategy is not used, the fundamental current of phase A has obvious phase lag and amplitude attenuation. In the power spectral density (PSD) distribution, the harmonic energy at this time is mainly concentrated in the 5th, 7th, 11th, 13th and 17th harmonics, and the 5th and 7th harmonics have the greatest interference; after using this strategy, the phase lag of the fundamental current is completely eliminated, the PSD distribution is more balanced, and the amplitude of each harmonic is greatly weakened. Figure 9 The waveform diagram shows the comparison of position estimation performance before and after using this strategy: when the strategy is not used, the estimated rotor position is distorted, and the estimated position is significantly lagging compared to the actual position signal obtained by the sensor. The center line of the rotor position estimation error fluctuation is not near the zero line, and the maximum position estimation error exceeds 0.3 rad; after using this strategy, the fluctuation of the rotor position estimation error is significantly reduced, and the magnitude does not exceed 0.1 rad.

Claims

1. A sensorless global delay compensation strategy for permanent magnet synchronous motors under square wave field weakening, characterized in that: Please follow the steps below to implement it: Step 1: Use the sliding mode observer in the rotating coordinate system as the position observation model to obtain the actual rotor position estimate Thus forming a speed loop; Step 2: Perform coordinate transformation based on the ideal and actual estimated rotor position angles to calculate the corresponding relationship between the ideal signal and the hysteresis signal of the quadrature and direct axis feedback currents; specifically: In step 2.1, without considering the calculation delay, the ideal stator three-phase current is shown in equation (3); Among them, i A is the phase A current, i B is the B phase current, i C is the C phase current, I m is the modulus of the current vector, θ ph is the ideal current vector angle; At this time, in position sensorless control, the change matrix R from the ABC phase to the dq coordinate system is shown in formula (4); in, is the ideal estimated rotor position angle; According to equation (4), coordinate transformation is performed on equation (3) to obtain the ideal quadrature and direct axis feedback current, as shown in equation (5); Where σ is the angle between the current vector I and the d-axis; Step 2.2, after considering the calculation delay, the current vector angle and the estimated rotor position angle have a The hysteresis angle is converted from equation (3) to equation (6), and from equation (4) to equation (7); Among them, I m_delay To consider the current vector modulus after delay, is the current phase lag angle; in, is the estimated current phase lag angle; According to equation (7), equation (6) is transformed into coordinates to obtain the actual quadrature and direct axis feedback current as shown in equation (8); Among them, i d_delay is the actual direct-axis feedback current; i q_delay is the actual quadrature-axis feedback current; In step 2.3, by combining equations (5) and (8), we can obtain the corresponding relationship between the ideal and actual signals of the quadrature and direct axis currents as shown in equation (9); Among them, M err is the amplitude error caused by the delay, which is written as formula (10) in the dq coordinate system; Step 3: Obtain the ideal and actual signals of the quadrature and direct axis feedback currents through a single q-axis current regulation position sensorless closed-loop structure; Step 4, performing amplitude error correction on the feedback current to calculate the estimated error of the phase lag angle; Step 5: Design an adaptive PI compensator to perform online PI adjustment on the estimated error of the phase lag angle, thereby eliminating the phase lag error; Step 6: Based on the voltage vector angle control of the single q-axis current regulator, the final voltage output command is obtained to achieve global delay compensation.

2. The sensorless global delay compensation strategy for permanent magnet synchronous motor under square wave field weakening according to claim 1 is characterized in that: In the step 1, specifically: Step 1.1, establish the dq sliding mode observer model, as shown in formula (1); in, and are the stator d-axis and q-axis current observation values, ω re is the rotor angular velocity, D is the differential operator, R s is the stator resistance, L d is the direct axis component of the stator inductance, L q is the quadrature-axis component of the stator inductance, u d is the direct axis component of the stator voltage, u q is the quadrature-axis component of the stator voltage; V d and V q are the control inputs of the sliding mode observer respectively; i d is the direct axis component of the stator current, i q is the quadrature-axis component of the stator current; Step 1.2: low-pass filter the PI module to get the estimated rotor angular velocity As shown in formula (2); Among them, K po is the proportional gain, K io is the integral gain; Estimating the angular velocity of the rotor By performing integration operation, the actual rotor position estimate can be obtained. This forms a speed loop.

3. The sensorless global delay compensation strategy for permanent magnet synchronous motor under square wave field weakening according to claim 2 is characterized in that: In the step 3, specifically: Step 3.1, for the three-phase current i A 、i B 、i C Sampling is performed, Clark transformation is performed, and the rotor position estimate obtained in step 1 is obtained. Perform Park transformation to obtain the actual quadrature and direct axis feedback current i dq_delay ; Step 3.2, obtain the ideal quadrature and direct axis feedback current signals under the field weakening constraint through the current trajectory planning link of the single q-axis current regulator field weakening control; Step 3.3, substitute the dq axis current command under the weak magnetic constraint into equations (9) and (10) to obtain equations (15) and (16); 4. The sensorless global delay compensation strategy for permanent magnet synchronous motor under square wave field weakening according to claim 3 is characterized in that: In the step 3.2, specifically: Step 3.2.1, calculate the preliminary d-axis current command through the MTPA control relationship As shown in formula (11); Among them, ψ f is the permanent magnet flux of the motor, is the q-axis current command; Step 3.2.2: Adjust the current command according to the field weakening control target; specifically: Step 3.2.2.1, the d-axis and q-axis feedforward voltages provided by the feedforward link and Calculate the feedback voltage amplitude As shown in formula (12); Step 3.2.2.2, set the voltage vector magnitude u max The weak magnetic compensation current Δi is obtained by PI adjustment of the error of the feedback voltage amplitude d,wkfd ; Step 3.2.2.3, the weak magnetic compensation current Δi output by the voltage closed loop d,wkfd , correct the d-axis current command to the optimal current trajectory, as shown in formula (13); Step 3.2.2.4, calculate the q-axis current command under the field weakening control constraint through the torque formula As shown in formula (14); 5. The sensorless global delay compensation strategy for permanent magnet synchronous motor under square wave field weakening according to claim 3 is characterized in that: In the step 4, specifically: Step 4.1, without considering the delay, eliminate the amplitude error of equation (15) to obtain equation (17); Among them, i d_com is the direct axis current after amplitude error correction, i q_com is the quadrature-axis current after amplitude error correction; Step 4.2, in square wave weak magnetic single q-axis current regulation, estimate the value Close to actual value hour, and Then equation (17) is transformed into equation (18); Based on formula (18), the estimation error of the phase lag angle is expressed as formula (19); 6. The sensorless global delay compensation strategy for permanent magnet synchronous motor under square wave field weakening according to claim 5 is characterized in that: In the step 5, specifically: Step 5.1, select the estimated error of the phase lag angle and its rate of change as the input of the parameter self-tuning module. The calculation rule in the self-tuning module is shown in formula (20); in, is the output of the parameter self-tuning module; k 11 、k 21 、k 12 and k 22 is the scale factor, b 11 、b 21 、b 12 and b 22 is the weighting factor; In step 5.2, the proportional and integral gains can be calculated using equation (21); Among them, ξ p is the proportional gain, ξ i is the integral gain; Step 5.3, the lag angle is obtained by performing online PI adjustment on the estimated error. When the estimated phase lag angle is compensated to the estimated value of the rotor position, the estimated rotor position angle is given by becomes As shown in formula (22); in, is the estimated value of the rotor position after compensation.

7. The sensorless global delay compensation strategy for a permanent magnet synchronous motor under square wave field weakening according to claim 6 is characterized in that: In step 6, specifically: Step 6.1, the estimated value of the rotor position after compensation obtained in step 5 Re-transform the three-phase current coordinates to obtain the stator current quadrature axis component i q , and for i q Perform amplitude error correction to obtain the q-axis current i after amplitude error correction q_corr ; Step 6.2, according to the q-axis current command under the weak magnetic control constraint obtained in step 3.2 and the q-axis current i after amplitude error correction obtained in step 6.1 q_corr , the voltage vector angle θ can be obtained from formula (23) VVA ; Step 6.3: Calculate the d-axis and q-axis command voltages output under the field-weakening control of the voltage vector angle of the single q-axis current regulator according to equation (24) to achieve global delay compensation for the single q-axis current regulation. Among them, u max is the voltage vector magnitude, u max =2u dc / π,u dc is the DC bus voltage.

Images