Sensorless control system and method for permanent magnet synchronous motor

Abstract

The application discloses a kind of permanent magnet synchronous motor sensorless control system and method, it is related to motor control technical field.The system includes positioning module, open loop strong drag operation module, sliding mode observer, state switching module and closed loop operation module.Positioning module locks rotor to initial reference angle;Open loop strong drag operation module uses incremental frequency control to drive motor start, and pre-calibrates the phase difference between open loop position and observation position;State switching module uses angle progressive alignment strategy to realize impactless transition when speed reaches threshold value;Closed loop operation module executes fuzzy adaptive PI control double closed loop vector control based on the position and speed estimation value output by sliding mode observer.The application effectively suppresses start step loss and switching oscillation, and improves the rotor speed and position estimation accuracy under variable load conditions.

Description

Technical Field

[0001] This invention relates to the field of motor control technology, and in particular to a sensorless control system and method for a permanent magnet synchronous motor. Background Technology

[0002] Permanent magnet synchronous motors (PMSMs) are widely used in various engineering fields such as industrial electrical drives, servo drives, new energy equipment, and home appliance frequency conversion due to their advantages such as high power density, high operating efficiency, low torque ripple, and excellent dynamic response. Sensorless control technology is one of the core technologies of PMSM drive systems. It estimates the rotor position and speed through algorithms without the need for position sensors, which can effectively reduce system hardware costs, improve equipment reliability, and broaden the applicability to complex working conditions. However, existing sensorless solutions using sliding mode observers (SMOs) estimate the rotor state based on the motor's back EMF. The accuracy of this estimation is highly dependent on the amplitude of the back EMF. During low-speed operation, the back EMF signal is weak and easily affected by noise, resulting in inaccurate observations and large position estimation deviations. Consequently, it cannot directly achieve smooth motor start-up and stable low-speed operation. The traditional segmented control scheme that combines incremental frequency (IF) open-loop start-up with SMO closed-loop has the drawbacks of large torque impact and obvious current oscillation when switching between open and closed loops, which can easily cause motor step loss and jitter. At the same time, in variable operating conditions where the load changes with the speed, it is difficult to balance the motor's starting tracking performance with high-speed steady-state control accuracy, and it cannot meet the control requirements of high-performance permanent magnet synchronous motor sensorless drive systems.

[0003] Compared to single open-loop or closed-loop sensorless control schemes, the segmented control scheme combining IF and SMO can effectively adapt to the operating requirements of motors and is currently the mainstream research direction in the field of sensorless control of permanent magnet synchronous motors. However, existing control schemes combining IF and SMO still have many shortcomings: First, the open-loop start-up phase often adopts an open-loop drive strategy with fixed torque current and fixed acceleration, which is prone to rotor speed lag and start-up step loss under load changes; Second, the switching process lacks a smooth phase alignment strategy, which is prone to torque impact and current overshoot due to sudden changes in coordinate transformation reference; Third, the speed loop control often uses a fixed-gain proportional-integral controller, which is difficult to simultaneously meet the requirements of low-speed observation noise suppression and high-speed steady-state accuracy control, and is prone to integral saturation and steady-state error exceeding the standard under load changes. Summary of the Invention

[0004] To address the above problems, this invention provides a sensorless control system and method for permanent magnet synchronous motors, aiming to solve the problems of easy step loss during open-loop start-up, large torque impact during open-loop / closed-loop switching, and poor steady-state accuracy under variable load conditions in the prior art. It achieves smooth start-up and shock-free switching of permanent magnet synchronous motors without sensors, effectively suppresses torque current oscillations, and improves the stability of full-speed-range control.

[0005] In a first aspect, the present invention provides a sensorless control system for a permanent magnet synchronous motor, comprising: The positioning module is used to apply fixed d-axis and q-axis current commands to the stator windings, lock the rotor to a preset initial reference angle, and switch to the open-loop strong drag operation module after a set time. Sliding mode observer is used to output rotor position and speed estimates in real time; The open-loop strong drag operation module is used to execute incremental frequency open-loop control to drive the motor to start, and generate the open-loop rotor position based on constant angular acceleration, while pre-calibrating the phase difference between the open-loop rotor position and the rotor position estimate. The state switching module is used to gradually eliminate the phase difference by employing an angle progressive alignment strategy when the rotor speed estimate reaches the switching threshold. The closed-loop operation module is used to perform closed-loop vector control of speed and current loop based on the rotor position and speed estimates output by the sliding mode observer.

[0006] Furthermore, the positioning module is specifically used for: Fixed d-axis and q-axis current commands are injected into the stator windings, and a fixed initial electrical angle is output as the coordinate transformation reference. The rotor is locked to the preset initial reference position through directional magnetization. The d-axis current command value is 0, and the q-axis current command value is a preset constant value, which is used to generate a constant electromagnetic torque to start the motor.

[0007] By precisely locking the rotor to the preset initial reference position through directional magnetization, the adverse effects of initial position deviation on subsequent start-up control are eliminated, avoiding start-up step loss or current surge caused by position error. At the same time, zero d-axis current allows all stator current to be used to generate electromagnetic torque, maximizing the torque-to-current ratio and improving the efficiency and response speed of the motor during the start-up phase.

[0008] Furthermore, the open-loop forced drag operation module includes: An angular velocity generation unit is used to generate linearly increasing angular velocity commands based on a preset constant angular acceleration. The integration calibration unit is used to integrate the angular velocity command to obtain the open-loop rotor position command, and to calculate in real time the phase difference between the open-loop rotor position and the rotor position estimate output by the sliding mode observer. The calculation formula is as follows: ; ; in, It is a constant angular acceleration; For the remainder function; This is the position of the open-loop rotor; This is the estimated rotor position. This represents the rotor position phase difference.

[0009] By generating a linearly increasing angular velocity command through constant angular acceleration, a continuous and non-jumping open-loop rotor position signal can be obtained after integration and remainder operation, ensuring the smoothness of the open-loop drive process. At the same time, the phase difference between the open-loop rotor position and the sliding mode observer estimate is calculated in real time, providing accurate phase pre-calibration information for subsequent state switching. This effectively suppresses torque shocks and current oscillations caused by phase abrupt changes during open-loop and closed-loop switching, improving the robustness of the startup phase.

[0010] Furthermore, the state switching module includes: The threshold detector is used to monitor the speed estimate output by the sliding mode observer. When the speed estimate exceeds the preset switching speed threshold, it determines that the switching condition has been met. A phase aligner, used to progressively correct the phase difference in small steps after a switching condition is met, includes the following steps: Calculate the rotor position phase difference Adjustment angle ,in It is a symbolic function; when Exceeding the phase difference threshold At that time, according to Correcting phase difference; The rotor position estimate used in the final closed-loop operation is .

[0011] The switching timing is automatically determined by speed estimation, avoiding the problem of manually setting the switching time not matching the actual working conditions; the rotor position phase difference is gradually corrected by small steps, rather than a large adjustment at once, thereby eliminating torque shock and current overshoot caused by sudden changes in coordinate transformation reference; the progressive alignment strategy realizes soft switching from open loop to closed loop, ensuring that the motor speed is stable and the current is smooth during the transition.

[0012] Furthermore, the closed-loop operation module includes: The speed loop adjustment unit includes a fuzzy adaptive PI controller, which dynamically adjusts the proportional coefficient and integral coefficient of the fuzzy adaptive PI controller according to the difference between the estimated speed value and the reference speed value and their rate of change, and outputs the current reference values ​​of the d-axis and q-axis. The current loop adjustment unit is used to generate reference voltages for the d-axis and q-axis using a current loop PI controller based on the current reference values ​​for the d-axis and q-axis.

[0013] The speed loop regulating unit can dynamically adjust the proportional and integral coefficients of the PI controller based on the difference between the estimated speed and the reference value and its rate of change, thus balancing the requirements of noise suppression at low speeds and steady-state accuracy control at high speeds. The current loop regulating unit independently controls the d-axis and q-axis voltages, improving the system's rapid response to load disturbances. The combination of the dual closed-loop structure and fuzzy adaptive control enables the motor to maintain good tracking performance and anti-interference capability under varying load and speed conditions.

[0014] Furthermore, the speed loop adjustment unit is specifically used for: The speed difference and its rate of change are fuzzified, and the proportional coefficient and integral coefficient are dynamically adjusted according to a predefined fuzzy rule table. The centroid method is used to resolve fuzziness and obtain the actual proportional coefficients and integral coefficients. The speed difference is input into the adjusted PI controller, and the q-axis current reference value is obtained after anti-saturation processing. The reference value of the d-axis current is obtained using the maximum torque-current ratio control method.

[0015] By using a pre-defined fuzzy rule table, the PI parameters can be intelligently adjusted according to the speed error and its rate of change, enabling the controller to maintain optimal dynamic response under different operating conditions. The centroid method for fuzzy resolution ensures the continuity and accuracy of the output parameters. Anti-saturation processing effectively prevents speed overshoot and oscillation caused by integral saturation. Maximum torque-current ratio control minimizes the stator current under the condition of outputting the same electromagnetic torque, improving the overall energy efficiency of the system.

[0016] Furthermore, the sliding mode observer includes: The mathematical model building unit is used to establish a continuous domain sliding mode observer based on the per-unit mathematical model of a permanent magnet synchronous motor in a two-phase stationary coordinate system. Discretization unit, used to discretize the continuous domain sliding mode observer using the first-order Euler method; The back electromotive force estimation unit is used to estimate the back electromotive force in the continuous domain sliding mode observer based on the sign function, and to perform low-pass filtering on the estimated back electromotive force. The phase-locked loop unit is used to extract rotor position and speed estimates from the filtered back electromotive force.

[0017] A continuous-domain sliding mode observer is established based on a per-unit mathematical model, which reduces the impact of motor parameter variations on observation accuracy and enhances the algorithm's versatility. The first-order Euler method is used for discretization, which facilitates efficient implementation in digital controllers. The back EMF is estimated by a sign function and combined with low-pass filtering, which effectively suppresses the high-frequency chattering phenomenon inherent in sliding mode control. The phase-locked loop unit can accurately extract the rotor position and speed from the filtered back EMF, maintaining high observation accuracy even in the low-speed region where the back EMF amplitude is low, providing reliable state feedback for sensorless control across the entire speed domain.

[0018] Furthermore, the phase-locked loop unit includes: An error phase detector is used to calculate the angle estimation error signal based on the filtered back electromotive force. A loop filter is used to perform proportional-integral control on the error signal and output an estimated value of the rotor's electric angular velocity. An integrator is used to integrate the estimated rotor electrical angular velocity to obtain the estimated rotor position.

[0019] The aforementioned phase-locked loop structure is insensitive to noise and harmonics in the back electromotive force, and can maintain high-precision locking of the rotor position during dynamic processes such as motor acceleration and deceleration and sudden load changes, which significantly improves the robustness and dynamic performance of the sensorless control system.

[0020] Secondly, the present invention provides a sensorless control method for a permanent magnet synchronous motor, comprising: S1, apply fixed d-axis and q-axis current commands to the stator windings, lock the rotor to the preset initial reference angle, and switch to the open-loop operation stage after a set time; S2, during the open-loop operation phase, executes incremental frequency open-loop control to drive the motor to start, generates the open-loop rotor position based on constant angular acceleration, and pre-calibrates the phase difference between the open-loop rotor position and the rotor position estimate output by the sliding mode observer; S3, when the velocity estimate output by the sliding mode observer reaches the switching threshold, the phase difference is gradually eliminated by an angle progressive alignment strategy. S4 executes closed-loop vector control of speed and current loop based on the rotor position and speed estimates output by the sliding mode observer.

[0021] Compared with the prior art, the beneficial effects of the present invention are: the sensorless control method combining incremental frequency open-loop start-up and sliding mode observer can effectively solve the technical problems of torque impact and current oscillation when switching between open and closed loops of permanent magnet synchronous motors, as well as the difficulty in balancing start-up performance and steady-state accuracy under variable load conditions, compared with the traditional single open-loop, single closed-loop or segmented control schemes without smooth switching. It also reduces the controller's dependence on mechanical position sensors and accurate motor models, and is more suitable for permanent magnet synchronous motor drive scenarios where the load changes with the speed and the operating conditions are complex and varied. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this drawing or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this drawing. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0023] Figure 1 This is a system architecture diagram of the present invention; Figure 2 proportionality coefficient The fuzzy rules represent the intent; Figure 3 Integral coefficient The fuzzy rules represent the intent; Figure 4 The simulation results for the per-unit speed are shown in the figure. Figure 5 A graph showing the per-unit estimation error of rotational speed; Figure 6 The image shows the simulation results of the rotor position. Figure 7 This is a diagram showing the rotor position estimation error. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be described and illustrated below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. All other embodiments obtained by those skilled in the art based on the embodiments provided by this invention without inventive effort are within the scope of protection of this invention.

[0025] This invention provides a sensorless control system for a permanent magnet synchronous motor, such as... Figure 1 As shown, the system includes a positioning module, an open-loop forced drag operation module, a state switching module, a closed-loop operation module, a sliding mode observer, and an SVPWM control module.

[0026] Before the motor starts, the system first executes the positioning module. This module applies fixed d-axis and q-axis current commands to the stator windings, where the d-axis current command value is... q-axis current command value , A preset constant value is used to generate a constant electromagnetic torque. Simultaneously, a fixed initial electrical angle is output. As a reference for coordinate transformation, the rotor is locked to the initial electrical angle through directional magnetization. After a preset time Then, the system switches to the open-loop forced drag operation module.

[0027] During system operation, the sliding mode observer runs continuously to estimate rotor position and velocity in real time. The construction steps of the sliding mode observer are as follows: First, the derivation of the permanent magnet synchronous motor in a two-phase stationary coordinate system is given. - Mathematical model in coordinate system: ; in, These are the named values ​​of stator resistance, d-axis inductance, and q-axis inductance, respectively. It is the named value of the rotor's electrical angular velocity; They are Named values ​​of shaft current; They are respectively Named values ​​of shaft voltages; They are The nominal value of the shaft back electromotive force satisfies the following formula: ; in, These are the named values ​​of the d-axis and q-axis currents, respectively. It is the named value of the rotor position; It is a permanent magnet flux chain; It is the time derivative operator.

[0028] Transform the above mathematical model into a per-unit value form: ; in, This is the gain coefficient of the voltage term in the current state equation, reflecting the contribution of voltage to the rate of change of current after standardization. ; It is the voltage base value. , It is the rated voltage; It is the base current value. , It is the rated current; This is the damping coefficient of the current itself in the state equation. Since the resistance consumes energy, this term is negative, representing the natural decay tendency of the current. ; The coefficients of the cross-coupling term describe the effect of the rotating electromotive force generated by the salient pole effect on the current dynamics. ; It is the time base value. , It is the base value of angular velocity. , The rated frequency; , They are respectively Per-unit value of shaft current; They are respectively Per-unit value of shaft voltage; They are The per-unit value of the shaft back electromotive force; It is the per-unit value of the rotor's electrical angular velocity; It is the derivative operator.

[0029] Based on the mathematical model in per-unit form described above, a mathematical model for a continuous-domain sliding mode observer is established: ; in, yes Per-unit value of the shaft current estimate; yes Per-unit value of the estimated back electromotive force of the shaft; It is the per-unit value of the estimated electrical angular velocity output by the phase-locked loop.

[0030] The mathematical model of the continuous-domain sliding mode observer is discretized using the first-order Euler method. Let the control period be... The discretized sliding mode observer expression is: ; in, These are the previous control cycle's Shaft reference voltage; G is the discretized voltage term coefficient, used to convert the per-unit voltage into a per-unit current increment; F is the discretized current self-feedback coefficient, reflecting the current decay caused by resistance; H is the discretized cross-coupling term coefficient, used to describe the effect of the rotating electromotive force generated by the salient pole effect on the current of the other axis.

[0031] Back electromotive force estimate Satisfy the following formula: ; in, It is a symbolic function; , The sliding mode gains for the α-axis and β-axis, respectively, determine the strength of the sliding mode observer's response to current estimation errors.

[0032] To eliminate high-frequency jitter caused by the sign function, a low-pass filter is applied to the estimated back electromotive force, resulting in the filtered value. Axis back electromotive force Then, a phase-locked loop (PLL) is used to extract the estimated values ​​of rotor position and speed from the filtered back electromotive force.

[0033] Specifically, firstly, an error phase detector is constructed, based on... The angle information included defines the error signal: ; in, The error signal output by the phase detector can be approximated as such when the angle estimation error is very small. ; This is the rotor position estimate currently output by the phase-locked loop.

[0034] Next, the error signal output by the phase detector... Input loop filter to obtain the electric angular velocity estimate. The loop filter uses a PI controller. Then... Integrate to obtain the updated rotor position estimate. .

[0035] The sliding mode observer and phase-locked loop operate continuously during the positioning, open-loop operation, state switching and closed-loop operation phases, and their outputs are only used for control during the closed-loop operation phase.

[0036] A constant angular acceleration is generated during the open-loop strong drag operation phase. ,according to Generate a linearly increasing angular velocity command, integrate the angular velocity command and take the remainder to obtain the open-loop rotor position. and ensure It lies between 0 and 2π, and the specific calculation formula is as follows: ; in, It is the upper limit speed of the threshold for open-loop forced drag operation; It is the total time of open-loop forced drag operation; This is the remainder function.

[0037] During the open-loop strong drag operation phase, the d-axis and q-axis current commands remain unchanged. , The current loop PI controller generates d-axis and q-axis reference voltage vectors, which are then sent to the SVPWM module after coordinate transformation to drive the motor.

[0038] When the electric angular velocity estimate output by the sliding mode observer Exceed At this time, the rotor position error needs to be gradually corrected to reduce the impact during open-loop switching. This function is achieved by the phase aligner in the state switching module.

[0039] Specifically, calculate the rotor position error. The angle adjusted each time is .

[0040] when Exceeding the threshold At that time, according to After correction, the final rotor position estimate used for closed-loop operation is... .

[0041] The above correction process is executed in each control cycle, gradually reducing the phase difference until convergence, thereby achieving a smooth, soft handover from open-loop to closed-loop. All modules use this process during state transitions and subsequent closed-loop operation. As an estimate of the rotor position.

[0042] After the switch is completed, the system enters the closed-loop operation module, which includes a speed loop regulation unit and a current loop regulation unit.

[0043] The speed loop control unit employs a fuzzy adaptive PI controller, which adjusts the proportional coefficient in the PI controller based on the difference between the speed reference value and the speed estimate. and integral coefficient .

[0044] Specifically, calculate the speed estimation error. ,in This is a reference value for rotational speed. This is an estimated value for the rotational speed. ; Calculate the rate of change of speed estimation error .

[0045] Will and They are divided into 8 membership degrees: {NB, NM, NS, NZ, PZ, PS, PM, PB}, corresponding to negative large, negative medium, negative small, zero negative, zero positive, positive small, positive medium, and positive large, respectively. Based on a pre-defined fuzzy rule table (e.g., ... Figure 2 , 3 As shown), dynamically adjust the scaling factor. and integral coefficient When the speed estimation error is greater than or equal to the first threshold, increase... To reduce error, and at the same time reduce To prevent oversaturation; when the speed estimation error is between the first and second thresholds, take a moderate value to prevent drastic speed changes and eliminate steady-state error; when the speed estimation error is close to zero, increase... and To suppress disturbances and eliminate steady-state error.

[0046] The fuzzy logic was solved using the centroid method. and The value of will Input a PI controller and apply anti-saturation processing to the output to obtain the q-axis current reference value. The reference value of the d-axis current is obtained using the maximum torque-current ratio control method. .

[0047] The current loop regulating unit will and As input, d-axis reference voltages are generated using d-axis and q-axis current loop PI controllers, respectively. and q-axis reference voltage .

[0048] The result The voltage in the two-phase stationary coordinate system is obtained by inverse Park transformation. The voltage command is fed into the SVPWM control module. The SVPWM module generates PWM signals based on the voltage command, driving the inverter to generate three-phase current, thereby controlling the smooth operation of the permanent magnet synchronous motor. Simultaneously, After a unit delay, the signal is fed back to the sliding mode observer for calculation in the next control cycle.

[0049] To verify the effectiveness of the sensorless control strategy proposed in this invention, simulation experiments were conducted under variable load conditions. The results are as follows: Figures 4 to 7 As shown. Among them, Figure 4 The simulation results are for the per-unit speed values. Figure 5This corresponds to the speed estimation error. Figure 6 The simulation results show the rotor position. Figure 7 The error represents the rotor position estimation. Simulation results show that the present invention can achieve smooth operation during startup, switching, and steady-state phases, with small speed and position estimation errors, verifying the feasibility of the system.

[0050] It should be noted that the present invention is not limited to the above-described embodiments. The above embodiments are merely examples, and any embodiments that have the same structure and perform the same effects as the technical concept within the scope of the present invention are included within the scope of the present invention. Furthermore, various modifications that can be conceived by those skilled in the art to the embodiments, and other ways of constructing by combining some of the constituent elements of the embodiments, without departing from the spirit of the present invention, are also included within the scope of the present invention.

Claims

1. A sensorless control system for a permanent magnet synchronous motor, characterized in that, include: The positioning module is used to apply fixed d-axis and q-axis current commands to the stator windings, lock the rotor to a preset initial reference angle, and switch to the open-loop strong drag operation module after a set time. Sliding mode observer is used to output rotor position and speed estimates in real time; The open-loop strong drag operation module is used to execute incremental frequency open-loop control to drive the motor to start, and generate the open-loop rotor position based on constant angular acceleration, while pre-calibrating the phase difference between the open-loop rotor position and the rotor position estimate. The state switching module is used to gradually eliminate the phase difference by employing an angle progressive alignment strategy when the rotor speed estimate reaches the switching threshold. The closed-loop operation module is used to perform closed-loop vector control of speed and current loop based on the rotor position and speed estimates output by the sliding mode observer.

2. The sensorless control system for a permanent magnet synchronous motor according to claim 1, characterized in that, The positioning module is specifically used for: A fixed d-axis current command and a q-axis current command are injected into the stator winding, and a fixed initial electrical angle is output as the coordinate transformation reference. The rotor is locked to the preset initial reference position by directional magnetization. The d-axis current command value is 0, and the q-axis current command value is a preset constant value, which is used to generate a constant electromagnetic torque to start the motor.

3. The sensorless control system for a permanent magnet synchronous motor according to claim 1, characterized in that, The open-loop forced drag operation module includes: An angular velocity generation unit is used to generate linearly increasing angular velocity commands based on a preset constant angular acceleration. The integration calibration unit is used to integrate the angular velocity command to obtain the open-loop rotor position command, and to calculate in real time the phase difference between the open-loop rotor position and the rotor position estimate output by the sliding mode observer. The calculation formula is as follows: ； ； in, It is a constant angular acceleration; For the remainder function; This is the position of the open-loop rotor; This is the estimated rotor position. This represents the rotor position phase difference.

4. The sensorless control system for a permanent magnet synchronous motor according to claim 1, characterized in that, The state switching module includes: The threshold detector is used to monitor the speed estimate output by the sliding mode observer. When the speed estimate exceeds the preset switching speed threshold, it determines that the switching condition has been met. A phase aligner, used to progressively correct the phase difference in small steps after a switching condition is met, includes the following steps: Calculate the rotor position phase difference Adjustment angle ,in It is a symbolic function; when Exceeding the phase difference threshold At that time, according to Correcting phase difference; The rotor position estimate used in the final closed-loop operation is .

5. The sensorless control system for a permanent magnet synchronous motor according to claim 1, characterized in that, The closed-loop operation module includes: The speed loop adjustment unit includes a fuzzy adaptive PI controller, which dynamically adjusts the proportional coefficient and integral coefficient of the fuzzy adaptive PI controller according to the difference between the estimated speed value and the reference speed value and their rate of change, and outputs the current reference values ​​of the d-axis and q-axis. The current loop adjustment unit is used to generate reference voltages for the d-axis and q-axis using a current loop PI controller based on the current reference values ​​for the d-axis and q-axis.

6. The sensorless control system for a permanent magnet synchronous motor according to claim 5, characterized in that, The speed loop adjustment unit is specifically used for: The speed difference and its rate of change are fuzzified, and the proportional coefficient and integral coefficient are dynamically adjusted according to a predefined fuzzy rule table. The centroid method is used to resolve fuzziness and obtain the actual proportional coefficients and integral coefficients. The speed difference is input into the adjusted PI controller, and the q-axis current reference value is obtained after anti-saturation processing. The reference value of the d-axis current is obtained using the maximum torque-current ratio control method.

7. The sensorless control system for a permanent magnet synchronous motor according to claim 1, characterized in that, The sliding mode observer includes: The mathematical model building unit is used to establish a continuous domain sliding mode observer based on the per-unit mathematical model of a permanent magnet synchronous motor in a two-phase stationary coordinate system. Discretization unit, used to discretize the continuous domain sliding mode observer using the first-order Euler method; The back electromotive force estimation unit is used to estimate the back electromotive force in the continuous domain sliding mode observer based on the sign function, and to perform low-pass filtering on the estimated back electromotive force. The phase-locked loop unit is used to extract rotor position and speed estimates from the filtered back electromotive force.

8. A sensorless control system for a permanent magnet synchronous motor according to claim 7, characterized in that, The phase-locked loop unit includes: An error phase detector is used to calculate the angle estimation error signal based on the filtered back electromotive force. A loop filter is used to perform proportional-integral control on the error signal and output an estimated value of the rotor's electric angular velocity. An integrator is used to integrate the estimated rotor electrical angular velocity to obtain the estimated rotor position.

9. A sensorless control method for a permanent magnet synchronous motor, characterized in that, include: S1, apply fixed d-axis and q-axis current commands to the stator windings, lock the rotor to the preset initial reference angle, and switch to the open-loop operation stage after a set time; S2, during the open-loop operation phase, executes incremental frequency open-loop control to drive the motor to start, generates the open-loop rotor position based on constant angular acceleration, and pre-calibrates the phase difference between the open-loop rotor position and the rotor position estimate output by the sliding mode observer; S3, when the velocity estimate output by the sliding mode observer reaches the switching threshold, the phase difference is gradually eliminated by an angle progressive alignment strategy. S4 executes closed-loop vector control of speed and current loop based on the rotor position and speed estimates output by the sliding mode observer.

Images