Deadbeat predictive current control method for permanent magnet synchronous motor based on sliding mode anti-disturbance

Through the sliding mode anti-disturbance-free beat prediction current control method, combined with the dual sliding mode observer and sliding mode controller, the high-order harmonic problems caused by resistance, inductance and magnetic flux disturbance in PMSM motor are solved, and the control is improved Rodability and response speed reduce inverter loss.

CN118367832BActive Publication Date: 2024-11-22WUHAN BUSINESS UNIV
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410143437.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-02-01
Publication Date
2024-11-22
Estimated Expiration
2044-02-01

AI Technical Summary

Technical Problem

In the prior art, the high-order harmonic problem of PMSM motors makes it difficult to effectively suppress resistance, inductance and magnetic flux disturbances, affecting control performance.

Method used

Using the sliding mode anti-disturbance-free beat prediction current control method, by building a sliding mode anti-disturbance-free beat prediction current vector control structure model, a dual sliding mode anti-disturbance-free beat prediction current observer and sliding mode controller are established, combining the index approach law and The constant-velocity approach law design is used to construct a predictive current controller, and large vectors are hidden in SVPWM modulation, and only small vectors and zero vectors are used for modulation.

Benefits of technology

Effectively suppress resistance, inductance and magnetic flux disturbances, improve system control robustness, reduce inverter switching losses, improve the stability of the internal ring of the current and the response speed of the external ring, and reduce the impact of load torque disturbances.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118367832B_ABST
    Figure CN118367832B_ABST
Patent Text Reader

Abstract

A method for predictive current control of a permanent magnet synchronous motor based on sliding mode anti-disturbance includes: the first step: building a permanent magnet synchronous motor sliding mode anti-disturbance predictive current vector control structure model; the second step: establishing an inner loop dq axis dual sliding mode anti-disturbance observer model based on the constant speed reaching law; the third step: establishing a sliding mode controller model based on the model speed outer loop; the fourth step: establishing a current inner loop model predictive current controller model based on the exponential reaching law; the fifth step: building an approximate two-level SVPWM deadbeat controller model based on the injection of virtual vectors. This design can not only effectively suppress the influence of parameter disturbances such as resistance, inductance, and flux linkage on the system, and improve the robustness of system control, but also the system responds quickly and effectively suppresses the influence of load torque disturbances.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a deadbeat predictive current control method for a permanent magnet synchronous motor based on sliding mode anti-disturbance, and is specifically applicable to a three-level NPC driven PMSM, and proposes a predictive current control method based on sliding mode anti-disturbance. Background Art

[0002] As the most representative type of AC motor, PMSM has many advantages such as simple structure, high torque-to-inertia ratio, small size, and high efficiency. It has broad application prospects in the fields of precision control and locomotive traction. In terms of control methods, the rotor flux oriented vector control (FOC) based on the PI regulator is widely used. The key lies in the control of the stator current amplitude and spatial position, and the speed and stator quadrature and direct axis currents are controlled in a double closed loop, respectively, and the system has good steady-state performance. However, the parameter selection of the PI regulator depends on the mathematical model of the motor. Since PMSM is a strongly coupled nonlinear system, the changes in parameters such as motor resistance R, inductance L and flux f have a great influence on the steady-state performance of the system. Therefore, PI control cannot directly meet the high-performance and high-precision control requirements of the motor.

[0003] The core idea of ​​PMSM model predictive control is to predict the state at future time according to the mathematical model of inverter and motor, as well as the state of motor at present time, compare with the given value, and select the optimal voltage vector to act on the motor. Compared with PI regulator, model predictive control eliminates the need for complex parameter setting and has better steady-state performance. Similar to PI regulator, inaccurate motor model or deviation of model parameters will lead to inaccurate selection of control voltage vector, resulting in deviation of system output current. As time goes by, the accumulated deviation of the system continues to increase, thus deteriorating the control performance of the system. As a strong robust control that is insensitive to parameters and external disturbances, sliding mode control has been a research hotspot of PMSM in recent years. Some control methods are a high-order sliding mode observer, which uses the nth-order differential of the observer output to observe the motor flux and resistance, but multiple differentials can easily form high-order harmonics, making it difficult to fully guarantee the accuracy of the observation results; some control methods use the current and voltage deviations of the d-axis and q-axis to construct a parameter identification model, design a suitable control rate, and ensure the convergence of the observed parameters; a fast Fourier transform-based motor parameter offline identification method uses the fundamental waves of the stator voltage and current to offline identify the stator d-axis and q-axis inductance, but the data obtained by this method cannot adapt to different working conditions of the motor; some use an extended Kalman filter to perform online identification of the flux f and inductance L to achieve the optimal estimation of the system in the sense of minimum mean square error. Summary of the invention

[0004] The purpose of the present invention is to overcome the problem in the prior art that multiple differentiations easily form high-order harmonics, and to provide a permanent magnet synchronous motor zero-beat predictive current control method based on sliding mode anti-disturbance that can effectively suppress disturbances such as resistance, inductance, and flux linkage.

[0005] To achieve the above objectives, the technical solution of the present invention is:

[0006] A deadbeat prediction current control method for a permanent magnet synchronous motor based on sliding mode anti-disturbance, the deadbeat prediction current control method is used to control an NPC structure three-level inverter circuit of a permanent magnet synchronous motor, and the deadbeat prediction current control method comprises the following steps:

[0007] Step 1: Build a permanent magnet synchronous motor sliding mode anti-disturbance zero-beat predictive current vector control structure model:

[0008]

[0009] In the above formula: u d is the d-axis component of the stator voltage, u q is the q-axis component of the stator voltage, i d is the d-axis component of the stator current, i q is the q-axis component of the stator current, R s is the stator winding phase resistance, ω e is the rotor electrical angle, ψ d is the d-axis flux component, ψ q is the q-axis magnetic flux component, L s is the stator inductance, p=d / dt is the differential operator, ψ f is the permanent magnet flux, T e is the electromagnetic torque, n p is the pole pair number, T m is the load torque converted to the shaft end, J is the moment of inertia of the mechanical load converted to the shaft end, ω m is the rotor mechanical angular velocity, B is the viscous friction coefficient;

[0010] Step 2: Establish the inner loop dq axis dual sliding mode anti-disturbance observer model based on the constant velocity reaching law, as follows:

[0011] D-axis observer:

[0012]

[0013] Where: is the d-axis current component at the next moment, is the d-axis current component at the current moment, i q (k) is the q-axis current component at the current moment, is the estimated value of the total disturbance on the d-axis at the current moment, u d(k) is the d-axis voltage component at the current moment;

[0014] Q-axis observer:

[0015]

[0016] Where: is the q-axis current component at the next moment, is the q-axis current component at the current moment, i d (k) is the d-axis current component at the current moment, u q (k) is the q-axis voltage component at the current moment, is the estimated value of the total disturbance on the q-axis at the current moment;

[0017] Step 3: Establish a sliding mode controller model based on the model speed outer loop, and design the sliding mode controller using the exponential reaching law method:

[0018]

[0019] Where: is the differential of the sliding surface function, is the rate of change of the difference between the given speed and the feedback speed, is the second-order differential of the difference between the given speed and the feedback speed, C2 is an arbitrary positive real number, m is an arbitrary positive real number, ξ s >0, k>0;

[0020] Step 4: Establish a current inner loop model based on the exponential reaching law to predict the current controller model.

[0021] Predicted current equation:

[0022]

[0023] Where: e(k) is the rotor electrical angular velocity at the current moment; by shifting the terms in the above equation, we can solve for u d (k) and u q (k) value:

[0024]

[0025] At this time, the current inner loop model prediction current controller model based on the exponential reaching law is completed;

[0026] Step 5: Construct an approximate two-level SVPWM zero-beat controller model based on injected virtual vectors. The output of the current controller model predicted by the current inner loop model in the exponential reaching law method is used as the input of the approximate two-level SVPWM zero-beat controller model based on injected virtual vectors. Using the improved SVPWM algorithm, the duty cycle of each voltage vector is reasonably allocated to effectively reduce the switching loss of the inverter.

[0027] In the first step, the permanent magnet synchronous motor sliding mode anti-disturbance zero-beat predictive current control system uses the d-axis given current vector i d * = 0 rotor flux oriented vector control, the system consists of a speed outer loop and a current inner loop. The speed outer loop provides the stator current q-axis component given value i through a sliding mode controller. q * ,i d * 、i q * The observed values ​​of the d and q axis currents output by the sliding mode anti-disturbance observer Together they serve as the input of the deadbeat predictive current controller, and the output can obtain the control voltage V d With V q After the voltage is transformed into the αβ stationary coordinate system, the improved SVPWM technology is used to control the three-level NPC inverter to supply power to the permanent magnet synchronous motor; the current inner loop is based on the sliding mode anti-disturbance observer, and the real-time values ​​of the d and q axis currents measured by the current transformer are i d 、i q , and the rotor electrical angle ω e Together they serve as the input of the sliding mode anti-disturbance observer, and the output As the predictive current controller input, the rotor electrical angle ω e It is obtained by derivation of the rotor position angle θ.

[0028] Step 2: Establish the inner loop dq axis dual sliding mode anti-disturbance observer model based on the constant velocity reaching law, assuming that the disturbance caused by the permanent magnet synchronous motor parameters is bounded and in a very short sampling period T sw The voltage inside is constant, and the voltage equations of the motor's d-axis and q-axis can be expressed as:

[0029]

[0030]

[0031] Where: f d is the disturbance caused by the d-axis parameter, f q is the disturbance caused by the q-axis parameter, F d f d The rate of change, F q f q The rate of change, △Ls is the stator inductance disturbance, △Rs is the stator resistance disturbance, Δψ f is the stator flux disturbance; i d and f d is the state variable, ω e is the rotor electrical angular velocity;

[0032] Taking the d-axis feedback current as the observation object, a sliding mode anti-disturbance observer is designed:

[0033]

[0034] Where: is the rate of change of the estimated d-axis current, is the rate of change of the d-axis disturbance estimate, is the d-axis current, is the estimated value of the d-axis disturbance, m is the sliding mode control gain (m>0), e id is the error between the d-axis observed current and the feedback current u(e id ) represents the sliding mode control rate of the d-axis;

[0035] Compensating the d-axis disturbance value into the voltage equation yields:

[0036]

[0037] Discretize the above formula and finally get the d-axis observer:

[0038]

[0039] Where: is the d-axis current component at the next moment, is the d-axis current component at the current moment, i q (k) is the q-axis current component at the current moment, is the estimated value of the total disturbance on the d-axis at the current moment, u d (k) is the d-axis voltage component at the current moment;

[0040] Similarly, taking the q-axis feedback current as the observation object, the sliding mode anti-disturbance observer is designed as follows:

[0041]

[0042] Where: is the rate of change of the estimated value of the q-axis current, is the rate of change of the estimated value of the q-axis disturbance, is the q-axis current, is the estimated value of the q-axis disturbance, n is the sliding mode control gain (n>0), e iq is the error between the q-axis observed current and the feedback current u(e iq ) is the q-axis sliding mode control rate;

[0043] Compensating the d-axis disturbance value into the voltage equation yields:

[0044]

[0045] Discretize the above formula to get the q-axis observer:

[0046]

[0047] Where: is the q-axis current component at the next moment, is the q-axis current component at the current moment, i d (k) is the d-axis current component at the current moment, u q (k) is the q-axis voltage component at the current moment, is the estimated value of the total disturbance on the q-axis at the current moment;

[0048] At this point, the inner loop dq axis dual sliding mode anti-disturbance observer model based on the constant velocity reaching law is completed.

[0049] Step 3: Establish a model of speed outer loop sliding mode controller based on the model. To speed up the sliding surface speed of the system motion point and reduce the chattering near the sliding surface, the speed outer loop sliding mode controller is designed based on the exponential approach law; define the controller state variables:

[0050]

[0051] Where: m * For a given rotor mechanical angular velocity, ω m The actual rotor mechanical angular velocity measured by the encoder is the feedback rotor mechanical angular velocity, e ω is the difference between the given rotor mechanical angular velocity and the feedback rotor mechanical angular velocity, is the rate of change of the difference between the given speed and the feedback speed, is the rate of change of a given speed;

[0052] The sliding mode controller is designed using the exponential reaching law:

[0053]

[0054] Where: is the differential of the sliding surface function, is the second-order differential of the difference between the given speed and the feedback speed, C2 is an arbitrary positive real number, m is an arbitrary positive real number, ξ s >0, k>0;

[0055] At this point, the model construction of the model speed outer loop sliding mode controller is completed.

[0056] Step 4: Establish a current inner loop model based on the exponential reaching law to predict the current controller model.

[0057] The current inner loop model based on the exponential reaching law is established to predict the current controller model, and the voltage equation of the permanent magnet synchronous motor in the synchronous rotating coordinate system is:

[0058] In a very small sampling period T sw In the above formula, the d and q axis currents are discretized to obtain:

[0059]

[0060] Use discrete equations to predict the current i at the next moment d (k+1) and i q (k+1), assuming that the d-axis current given value is i d * =i d (k+1), q-axis current given value i q * =i q (k+1); with d and q axis stator voltage given value u d 、u q is the control rate, and the predicted current equation is obtained:

[0061]

[0062] Where: e(k) is the rotor electrical angular velocity at the current moment; by shifting the terms in the above equation, we can solve for u d (k) and u q (k) value:

[0063]

[0064] At this point, the current inner loop model predictive current controller model based on the exponential reaching law is completed.

[0065] Step 5: Construct an approximate two-level SVPWM deadbeat controller model based on injected virtual vector;

[0066] Sector division: Divide the input voltage space vector into large and small sectors, and improve the SVPWM algorithm: Based on the principle of hiding the medium vector and large vector, and only using the zero vector, small vector and virtual vector for modulation, the area where the voltage space vector passes is divided into 12 asymmetric sectors;

[0067] Formulate the switching sequence: Arrange the switching sequence of each small sector inverter to form a switching sequence chart. In order to reduce the stress of the switching devices, the 7-segment symmetrical voltage vector and the 9-segment symmetrical voltage vector are switched to each other. The variable sampling period is used to control the target space voltage vector V ref The sector number it falls into is used to formulate the action sequence table of the basic voltage vector;

[0068] Vector action time calculation: determine the action time of each switching vector, form the SVPWM control model, and define the zero vector as V s0 , the small vector on the left side of the large sector is V s1 , the small vector on the right side of the large sector is V s2 When the 7-segment symmetrical voltage vector control is used, a two-level algorithm is used, using V s1 、V s2 、V s0 The target voltage vector V ref , V s1 、V s2 、V s0 The vector action time corresponds to T1, T2, and T0 respectively; when the 9-segment symmetrical voltage vector control is adopted, an approximate two-level algorithm of injecting virtual vector is adopted, and the virtual vector is further defined as V s3 , V s1 、V s2 Vector is used to synthesize virtual vector V s3 , use V s0 、V s1 、V s2 、V s3 The target voltage vector V ref , V s1 、V s2 、V s0 The vector action time corresponds to (T1′+T sw ′), T2′, T0′; T sw For a switching cycle, according to the vector synthesis law of different small sectors, the action time of each vector in each small sector is calculated respectively; then, according to the basic vector action time, the combined switch control pulse corresponding to each sector is obtained;

[0069] Finally, according to the switching state and trigger time of each sector, the trigger time table of the switch devices in each sector is sorted out to form the SVPWM control model. The above control model is used to control the permanent magnet synchronous motor NPC inverter circuit.

[0070] A deadbeat predictive current control system for a permanent magnet synchronous motor based on sliding mode anti-disturbance, wherein the deadbeat predictive current control system for a permanent magnet synchronous motor based on sliding mode anti-disturbance is a deadbeat predictive current control model constructed by the deadbeat predictive current control method for a permanent magnet synchronous motor based on sliding mode anti-disturbance;

[0071] The control system comprises:

[0072] Deadbeat predictive current vector controller: a mathematical model for the synchronous motor sliding mode anti-disturbance deadbeat predictive current vector control;

[0073] Inner loop dq axis dual sliding mode anti-disturbance observer: Algorithm for executing the inner loop dq axis dual sliding mode anti-disturbance observer;

[0074] Speed ​​loop sliding mode controller: used to execute the model speed outer loop sliding mode control algorithm;

[0075] Predictive current controller: used to execute the current inner loop model algorithm based on the exponential reaching law method;

[0076] Approximate two-level SVPWM deadbeat controller based on injected virtual vector: used to implement the improved SVPWM algorithm.

[0077] A deadbeat predictive current control device for a permanent magnet synchronous motor based on sliding mode anti-disturbance, comprising a memory and a processor, wherein the memory is used to store a computer program code and transmit the computer program code to the processor;

[0078] The processor is used to execute the permanent magnet synchronous motor zero-beat predictive current control method based on sliding mode anti-disturbance according to the instructions in the computer program code.

[0079] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for zero-beat prediction current control of a permanent magnet synchronous motor based on sliding mode anti-disturbance is implemented.

[0080] Compared with the prior art, the present invention has the following beneficial effects:

[0081] 1. The present invention relates to a deadbeat predictive current control method for a permanent magnet synchronous motor based on sliding mode anti-disturbance. The current inner loop adopts a constant velocity reaching law sliding mode observer to estimate the total parameter disturbance suffered by the system, and implements real-time feedforward compensation for the model. This method can effectively suppress the influence of parameter disturbances such as resistance, inductance, and flux linkage on the system, and improve the robustness of system control.

[0082] 2. The present invention relates to a method for zero-beat predictive current control of a permanent magnet synchronous motor based on sliding mode anti-disturbance. The speed outer loop adopts a sliding mode controller based on the exponential reaching law. Compared with the traditional speed outer loop control based on the PI controller, the system responds quickly and can effectively suppress the influence of load torque disturbance, thereby enhancing robustness.

[0083] 3. The present invention relates to a deadbeat predictive current control method for a permanent magnet synchronous motor based on sliding mode anti-disturbance, which hides large vectors and medium vectors in the SVPWM modulation process and only involves small vector and zero vector synthesis operations. The method has a good inhibitory effect on the deviation of the midpoint potential on the DC side of the inverter, and can effectively reduce the harmonic component of the stator current of the permanent magnet synchronous motor.

[0084] 4. The present invention provides a permanent magnet synchronous motor zero-beat predictive current control system based on sliding mode anti-disturbance, comprising: a zero-beat predictive current vector controller: a mathematical model of the synchronous motor sliding mode anti-disturbance zero-beat predictive current vector control; an inner loop dq axis dual sliding mode anti-disturbance observer: used to execute the algorithm of the inner loop dq axis dual sliding mode anti-disturbance observer; a speed loop sliding mode controller: used to execute the model speed outer loop sliding mode control algorithm; a predictive current controller: used to execute the current inner loop model algorithm based on the exponential reaching law method; an approximate two-level SVPWM zero-beat controller based on injected virtual vectors: used to execute the improved SVPWM algorithm.

[0085] 5. A deadbeat predictive current control device for a permanent magnet synchronous motor based on sliding mode anti-disturbance of the present invention comprises a processor and a memory, the memory is used to store computer program code, and transmit the computer program code to the processor, and the processor is used to execute the deadbeat predictive current control method for a permanent magnet synchronous motor based on sliding mode anti-disturbance provided in any of the above technical solutions according to the instructions in the computer program code. Therefore, the device also includes all the beneficial effects of the deadbeat predictive current control method for a permanent magnet synchronous motor based on sliding mode anti-disturbance provided in any of the above technical solutions, which will not be repeated here.

[0086] 6. A computer-readable storage medium of the present invention stores a computer program thereon, and when the computer program is executed by a processor, the steps of the method for predictive current control of a permanent magnet synchronous motor based on sliding mode anti-disturbance as provided in any of the above technical solutions are implemented. Therefore, the computer-readable storage medium also includes all the beneficial effects of the method for predictive current control of a permanent magnet synchronous motor based on sliding mode anti-disturbance as provided in any of the above technical solutions, which will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 It is a structural schematic diagram of the system of the present invention.

[0088] Figure 2 yes Figure 1 Principle block diagram of the inner ring dq axis dual sliding mode anti-disturbance observer.

[0089] Figure 3 yes Figure 1 Principle block diagram of the medium speed loop sliding mode controller.

[0090] Figure 4 yes Figure 1 Figure 2. Schematic diagram of the predictive current controller.

[0091] Figure 5 yes Figure 1 Flowchart of the improved SVPWM algorithm implementation.

[0092] Figure 6This is a diagram of the sector division method of the improved SVPWM algorithm.

[0093] Figure 7 This is a workflow diagram of the system within a single sampling period in Example 2. DETAILED DESCRIPTION

[0094] The present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0095] Embodiment 1:

[0096] See also Figures 1 to 6 , a deadbeat predictive current control method for a permanent magnet synchronous motor based on sliding mode anti-disturbance, comprising the following steps:

[0097] Step 1: Build a permanent magnet synchronous motor sliding mode anti-disturbance zero-beat predictive current vector control structure model;

[0098] The mathematical model and structural model of the sliding mode anti-disturbance zero-beat predictive current vector control of the permanent magnet synchronous motor are constructed. The mathematical model of the permanent magnet synchronous motor is shown in equations (1) to (4):

[0099]

[0100] In the above formula: u d is the d-axis component of the stator voltage; u q is the q-axis component of the stator voltage; i d is the d-axis component of the stator current; i q is the q-axis component of the stator current; R s is the stator winding phase resistance; ω e is the rotor electrical angle; ψ d is the d-axis magnetic flux component; ψ q is the q-axis magnetic flux component; L s is the stator inductance; p=d / dt is the differential operator; ψ f is the permanent magnet flux; T e is the electromagnetic torque; n p is the pole pair number; T m is the load torque converted to the shaft end; J is the moment of inertia of the mechanical load converted to the shaft end; ω m is the rotor mechanical angular velocity; B is the viscous friction coefficient;

[0101] The sliding mode anti-disturbance zero-beat predictive current control system of permanent magnet synchronous motor is based on the given current vector i of the d-axis. d * = 0 rotor flux oriented vector control, the system consists of a speed outer loop and a current inner loop. The speed outer loop provides the stator current q-axis component given value i through a sliding mode controller. q * ,id * 、i q * The observed values ​​of the d and q axis currents output by the sliding mode anti-disturbance observer Together they serve as the input of the deadbeat predictive current controller, and the output can obtain the control voltage V d With V q After the voltage is transformed into the αβ stationary coordinate system, the improved SVPWM technology is used to control the three-level NPC inverter to supply power to the permanent magnet synchronous motor; the current inner loop is based on the sliding mode anti-disturbance observer, and the real-time values ​​of the d and q axis currents measured by the current transformer are i d 、i q , and the rotor electrical angle ω e Together they serve as the input of the sliding mode anti-disturbance observer, and the output As the predictive current controller input, the rotor electrical angle ω e Obtained by derivation of the rotor position angle θ;

[0102] Step 2: Establish an inner loop dq axis dual sliding mode anti-disturbance observer model based on the constant speed reaching law, and use the saturation function sat(s) to replace the traditional sgn(s) function to effectively reduce the observer jitter; establish an inner loop dq axis dual sliding mode anti-disturbance observer model based on the constant speed reaching law, assuming that the disturbance caused by the permanent magnet synchronous motor parameters is bounded and in a very short sampling period T sw The voltage inside is constant, and the voltage equations of the motor's d-axis and q-axis can be expressed as:

[0103]

[0104] Where: f d is the disturbance caused by the d-axis parameter, f q is the disturbance caused by the q-axis parameter, F d f d The rate of change, F q f q The rate of change, △Ls is the stator inductance disturbance, △Rs is the stator resistance disturbance, Δψ f is the stator flux disturbance; i d and f d As the state variable, we get the state space function:

[0105]

[0106] Taking the d-axis feedback current as the observation object, a sliding mode anti-disturbance observer is designed:

[0107]

[0108] Where: is the rate of change of the estimated d-axis current, is the rate of change of the d-axis disturbance estimate, is the d-axis current, is the estimated value of the d-axis disturbance, m is the sliding mode control gain (m>0), e id is the error between the d-axis observed current and the feedback current u(e id ) represents the d-axis sliding mode control rate, and the observer output y is the d-axis observed current value:

[0109]

[0110] Subtracting equation (7) from equation (8) yields the error equation:

[0111]

[0112] Where: is the error change rate between the d-axis observed current and the feedback current, is the error change rate between the estimated disturbance and the actual disturbance on the d-axis, e fd is the error between the estimated disturbance and the actual disturbance of the d-axis Design a reasonable control rate u(e id ), ensuring the d-axis current observation value Under the premise of the existence of system disturbance, it can also converge to the feedback current i in a limited time. d ; To ensure that the system state is located on the sliding surface at the initial moment, the nonlinear integral sliding surface function S is selected:

[0113]

[0114] Where c is a positive real number. To further weaken the high-frequency chattering caused by the switching of the sliding mode surface, the saturation function sat(s) is used to replace the traditional sgn(s) function:

[0115]

[0116] Where ρ is the function boundary value, which is a positive real number. The sliding mode control function is designed by the constant velocity reaching law:

[0117]

[0118] In the formula: The reaching law parameter ξ d is a positive real number, is the differential of the sliding surface function; update the above formula to obtain the control rate u(e id ):

[0119]

[0120] Combining equations (8) and (14), we can get the d-axis disturbance value:

[0121]

[0122] Define the Lyapunov function V s , take the derivative of the function, and connect equation (13) and equation (14) to get:

[0123]

[0124] Where: V s The derivative of

[0125] From the above formula, we can get the control rate u(e id ) value complies with Lyapunov's second stability criterion, and the observer system is asymptotically stable over a large range; the d-axis disturbance value is compensated into the voltage equation, and equation (5) is rewritten to obtain:

[0126]

[0127] By discretizing equation (18), we finally get the d-axis observer:

[0128]

[0129] Where: is the d-axis current component at the next moment, is the d-axis current component at the current moment, i q (k) is the q-axis current component at the current moment, is the estimated value of the total disturbance on the d-axis at the current moment, u d (k) is the d-axis voltage component at the current moment;

[0130] Similarly, taking the q-axis feedback current as the observation object, the sliding mode anti-disturbance observer is designed as follows:

[0131]

[0132] Where: is the rate of change of the estimated value of the q-axis current, is the rate of change of the estimated value of the q-axis disturbance, is the q-axis current, is the estimated value of the q-axis disturbance, n is the sliding mode control gain (n>0), e iq is the error between the q-axis observed current and the feedback current u(e iq ) is the q-axis sliding mode control rate; the control rate u(e iq ) is expressed as:

[0133]

[0134] Where: n, C1 are positive real numbers, and the reaching law parameter ξ q is a positive real number; the q-axis perturbation value is:

[0135]

[0136] Rewrite equation (6) and discretize it:

[0137]

[0138] Get the q-axis observer:

[0139]

[0140] Where: is the q-axis current component at the next moment, is the q-axis current component at the current moment, i d (k) is the d-axis current component at the current moment, u q (k) is the q-axis voltage component at the current moment, is the estimated value of the total disturbance on the q-axis at the current moment;

[0141] At this point, the inner ring dq axis dual sliding mode anti-disturbance observer model based on the constant velocity reaching law is completed;

[0142] Step 3: Establish a model of speed outer loop sliding mode controller based on the model. To speed up the sliding surface speed of the system motion point and reduce the chattering near the sliding surface, the speed outer loop sliding mode controller is designed based on the exponential approach law; define the controller state variables:

[0143]

[0144] Where: m * For a given rotor mechanical angular velocity, ω m The actual rotor mechanical angular velocity measured by the encoder is the feedback rotor mechanical angular velocity, e ω is the difference between the given rotor mechanical angular velocity and the feedback rotor mechanical angular velocity, is the rate of change of the difference between the given speed and the feedback speed, is the rate of change of a given speed;

[0145] Since the inner current loop uses i d * =0 control, the permanent magnet synchronous motor torque and motion equation can be changed to the following formula:

[0146]

[0147] Where: is the second-order differential of the difference between the given speed and the feedback speed, is the second-order differential of a given velocity, n p is the pole pair number, J is the moment of inertia, B is the viscous friction coefficient, for

[0148] Differential of q-axis current, T m is the load torque.

[0149] Take iq as the control rate and select the linear sliding surface function S1, as shown in the following formula:

[0150]

[0151] The sliding mode controller is designed using the exponential reaching law:

[0152]

[0153] Where: is the differential of the sliding surface function, is the second-order differential of the difference between the given speed and the feedback speed, C2 is an arbitrary positive real number, m is an arbitrary positive real number, ξ s >0, k>0;

[0154] To prove the stability of the sliding mode controller, we define the Lyapunov function At this time there are:

[0155]

[0156] The above formula satisfies the large-scale asymptotic stability condition, and it can be concluded that the sliding mode controller with exponential reaching law is stable. At this time, the model of the sliding mode controller based on the model speed outer loop is completed.

[0157] Discretizing equation (29), we can get the controller principle block diagram as follows: Figure 4 shown.

[0158] Step 4: Establish a current inner loop model predictive current controller model based on the exponential reaching law, construct the Lyapunov function, prove the stability of the controller, and build a model framework based on Simulink;

[0159] The current inner loop model based on the exponential reaching law is established to predict the current controller model, and the voltage equation of the permanent magnet synchronous motor in the synchronous rotating coordinate system is:

[0160]

[0161] In a very small sampling period T sw In the above formula, the d and q axis currents are discretized to obtain:

[0162]

[0163] Use discrete equations to predict the current i at the next moment d (k+1) and i q (k+1), assuming that the d-axis current given value is i d * =i d (k+1), q-axis current given value i q * =i q (k+1); with d and q axis stator voltage given value u d 、u q is the control rate, and the predicted current equation is obtained:

[0164]

[0165] Where: e(k) is the rotor electrical angular velocity at the current moment; by shifting the terms in the above equation, we can solve for u d (k) and u q (k) value:

[0166]

[0167] At this point, the current inner loop model prediction current controller model based on the exponential approach is completed; the principle block diagram is as follows Figure 5 shown.

[0168] Step 5: Build an approximate two-level SVPWM deadbeat controller model based on injected virtual vectors, use the improved SVPWM algorithm to reasonably allocate the duty cycle of each voltage vector, effectively reduce the switching loss of the inverter, and build a model framework based on Simulink;

[0169] Sector division: Divide the input voltage space vector into large and small sectors, and improve the SVPWM algorithm: Based on the principle of hiding the medium vector and large vector, and only using the zero vector, small vector and virtual vector for modulation, the area where the voltage space vector passes is divided into 12 asymmetric sectors;

[0170] Formulate the switching sequence: Arrange the switching sequence of each small sector inverter to form a switching sequence chart. In order to reduce the stress of the switching devices, the 7-segment symmetrical voltage vector and the 9-segment symmetrical voltage vector are switched to each other. The variable sampling period is used to control the target space voltage vector V ref The sector number it falls into is used to formulate the action sequence table of the basic voltage vector;

[0171] Vector action time calculation: determine the action time of each switching vector, form the SVPWM control model, and define the zero vector as Vs0 , the small vector on the left side of the large sector is V s1 , the small vector on the right side of the large sector is V s2 When the 7-segment symmetrical voltage vector control is used, a two-level algorithm is used, using V s1 、V s2 、V s0 The target voltage vector V ref , V s1 、V s2 、V s0 The vector action time corresponds to T1, T2, and T0 respectively; when the 9-segment symmetrical voltage vector control is adopted, an approximate two-level algorithm of injecting virtual vector is adopted, and the virtual vector is further defined as V s3 , V s1 、V s2 Vector is used to synthesize virtual vector V s3 , use V s0 、V s1 、V s2 、V s3 The target voltage vector V ref , V s1 、V s2 、V s0 The vector action time corresponds to (T1′+T sw ′), T2′, T0′; T sw For a switching cycle, according to the vector synthesis law of different small sectors, the action time of each vector in each small sector is calculated respectively; then, according to the basic vector action time, the combined switch control pulse corresponding to each sector is obtained;

[0172] Finally, according to the switching state and trigger time of each sector, the trigger time table of the switch devices in each sector is sorted out to form the SVPWM control model. The above control model is used to control the permanent magnet synchronous motor NPC inverter circuit.

[0173] Simulation verification, using simulation software to establish the algorithm simulation platform, the obtained SVPWM control model is simulated and calculated, and its waveform quality and harmonic level are compared.

[0174] The specific division method of the sector division is as follows:

[0175] The input voltage space vector is divided into large and small sectors, and the SVPWM algorithm is improved: based on the principle of hiding the medium vector and large vector and only using the zero vector, small vector and virtual vector for modulation, the area where the voltage space vector passes is divided into 12 asymmetric sectors;

[0176] In the mutually perpendicular α and β vector coordinate systems, the α and β vector coordinate systems are equally divided into 6 large sectors, the 3rd large sector is the one with an angle θ of 0 to 60 degrees with the α axis; the 1st large sector is the one with an angle θ of 60 to 120 degrees with the α axis; the 5th large sector is the one with an angle θ of 120 to 180 degrees with the α axis; the 4th large sector is the one with an angle θ of 180 to 240 degrees with the α axis; the 6th large sector is the one with an angle θ of 240 to 300 degrees with the α axis; the 2nd large sector is the one with an angle θ of 300 to 360 degrees with the α axis;

[0177] The medium vector and the large vector are hidden, and modulation is performed using the zero vector, the small vector and the virtual vector. According to the above modulation method, each sector only needs to be divided into two small sectors, and the division rule is as follows:

[0178] The zero vector corresponds to the coordinate origin. The area completely modulated by the small vector in each large sector belongs to the small sector No. 1, and the area modulated by the small vector and the virtual vector belongs to the small sector No. 2.

[0179] The target voltage vector is V ref , target voltage vector V ref The angle between the α axis and the voltage vector is determined to be V according to the angle θ. ref The number of the large sector where the target voltage vector V ref The angle θ is converted to the No. 3 large sector to obtain the converted angle θ':

[0180]

[0181] It can be concluded that according to V α and V β The value of determines the rule for determining the small sector number, and the determination rules are as follows:

[0182] satisfy Belongs to small sector No. 1;

[0183] satisfy Belongs to small sector No. 2.

[0184] The target voltage vector V ref By combining the large sector number with the small sector number, the target voltage vector V ref The full sector number.

[0185] Formulate the switching sequence: Arrange the switching sequence of each small sector inverter to form a switching sequence chart. In order to reduce the stress of the switching devices, a variable sampling period is used to calculate the switching sequence according to the target space voltage vector V refThe sector number falls into, and the action sequence table of the basic voltage vector is formulated; Different from the traditional three-level SVPWM algorithm, in order to synthesize the target flux of a switching cycle, this algorithm adopts the method of switching between 7-segment symmetrical voltage vector and 9-segment symmetrical voltage vector for control. ref Falling into small sector 1, it is controlled according to the 7-segment symmetrical voltage vector algorithm; if V ref If it falls into small sector No. 2, according to the 9-segment symmetrical voltage vector algorithm, the sampling period is shortened and the sampling frequency is doubled; according to the above rules, the voltage vector action sequence and timetable of small sectors No. 1 and No. 2 are formulated respectively;

[0186] Table 1: Space voltage vector action sequence table of small sector 1

[0187]

[0188] Table 2: Space voltage vector action sequence table of small sector 2

[0189]

[0190] Table 1 specifies the order of action of the basic voltage vectors.

[0191] The sampling method with variable sampling period, if V ref If it falls into small sector 1, the system single scanning cycle is maintained at T sw ; If V ref If it falls into small sector 2, the system single scanning cycle is T sw ′, and T sw ′=0.5T sw , correspondingly, T0′=0.5T0; T1′=0.5T1; T2′=0.5T2, then: T0′+T1′+T2′+0.5T sw ′+0.5T sw ′ = Tsw;

[0192] When the target voltage vector falls into the small sector 1, at T sw The voltage modulation is performed only once during the time; when the target voltage vector falls into the small sector 2, at T sw During the time, the controller automatically allocates two time periods of 0.5T. sw The scanning period T sw ', performing voltage modulation twice, essentially increasing the switching frequency of the power electronic devices.

[0193] In the vector action time calculation, if the target voltage vector falls into small sector 1, a two-level algorithm is used, and the zero vector is defined as V s0 , the small vector on the left side of the large sector is V s1 , the small vector on the right side of the large sector is Vs2 , using only V s1 、V s2 、V s0 Synthesized target voltage vector V ref , where V s1 、V s2 、V s0 The vector action time corresponds to T1, T2, and T0 respectively, then:

[0194]

[0195] If the target voltage vector falls into the small sector 2, the approximate two-level algorithm of injecting virtual vector is adopted, and the virtual vector is further defined as V s3 , V s1 、V s2 Vector is used to synthesize virtual vector V s3 , use V s0 、V s1 、V s2 、V s3 The target voltage vector V ref , V s1 、V s2 、V s0 The vector action time corresponds to (T1′+T sw ′), T2′, T0′, then:

[0196]

[0197] Among them, the first 0.5T sw The basic vectors V0, V1, and V2 are used to synthesize the target flux 0.5V3·T corresponding to the virtual vector V3. sw , this magnetic flux is the first branch of the final target magnetic flux, which is expressed by the formula:

[0198]

[0199] The second 0.5T sw In the example, the basic vector V1 is introduced again, and the action time of this vector is 0.5T. sw , forming the second branch of the final target magnetic chain;

[0200] The final target magnetic flux expression is:

[0201] 0.5V3·T sw +0.5V1·T sw =V ref ·T sw (39);

[0202] According to the vector synthesis law of different small sectors, the action time of each vector in each small sector is calculated respectively: define the intermediate variable The values ​​of T0, T1, T2 and T0′, T1′, T2′ are calculated respectively:

[0203]

[0204]

[0205] Combine the switch action sequence table and the formula of each vector action time to solve the basic vector action time of each sector, and then obtain the combined switch control pulse situation corresponding to each sector based on the basic vector action time;

[0206] The combined switch control pulse conditions corresponding to each sector are obtained by using the obtained information, and modulation is performed by switching between 7-segment symmetrical voltage vectors and 9-segment symmetrical voltage vectors.

[0207] In V ref When falling into small sector 1, the modulation method of 7-segment symmetrical voltage vector switching time is adopted: the switch trigger variables A, B, and C are defined as follows:

[0208]

[0209] In V ref When falling into small sector 2, the modulation method of 9-segment symmetrical voltage vector switching time is adopted:

[0210] Introduce the basic voltage vector V1 and act In a T sw 'The counter is used to generate an isosceles triangle wave and compare it with a constant value Compare to determine the switch moment and define the switch trigger variables D, E, and F:

[0211]

[0212] Then introduce the basic voltage vector V1 and act

[0213] According to the switch state and trigger time of each sector, the trigger time table of the switch devices in each sector is sorted out to form the control model of SVPWM.

[0214] Embodiment 2:

[0215] A deadbeat predictive current control system for a permanent magnet synchronous motor based on sliding mode anti-disturbance, characterized in that:

[0216] The deadbeat predictive current control system for a permanent magnet synchronous motor based on sliding mode anti-disturbance is a deadbeat predictive current control model constructed by the deadbeat predictive current control method for a permanent magnet synchronous motor based on sliding mode anti-disturbance described in Example 1;

[0217] The control system comprises:

[0218] Deadbeat predictive current vector controller: a mathematical model for the synchronous motor sliding mode anti-disturbance deadbeat predictive current vector control;

[0219] Inner loop dq axis dual sliding mode anti-disturbance observer: Algorithm for executing the inner loop dq axis dual sliding mode anti-disturbance observer;

[0220] Outer loop sliding mode controller: used to execute the model speed outer loop sliding mode control algorithm;

[0221] Predictive current controller: used to execute the current inner loop model algorithm based on the exponential reaching law method;

[0222] Approximate two-level SVPWM deadbeat controller based on injected virtual vector: used to implement the improved SVPWM algorithm.

[0223] PMSM parameters and simulation environment

[0224]

[0225] The SMO+MPC method is used, and the structural block diagram is shown in Figure 1 As shown, the implementation process is as follows Figure 7 shown.

[0226] The simulation time is set to 1s, the motor given speed is 1000r / min, the motor starts with 0 load, and the torque Tm=10Nm is suddenly added at t=0.2s. The load torque is added at 0.2s. The PMSM speed drops to about 950r / min under both controllers and rises to 1000r / min in about 0.25s. Both controllers can respond quickly to changes in load torque, but compared with the SMO+MPC controller, the PI controller has no obvious jitter, the torque waveform is more stable, and the pulsation is smaller.

[0227] Comparison results of motor torque and speed response under two controllers. The output ripple of PMSM torque and speed based on PI controller is greater than that of SMO+MPC controller. But in general, the change of stator resistance Rs has little effect on the control effect.

[0228] Under the same switching frequency, the total harmonic distortion THD based on PI controller is 13.98%; the total harmonic distortion THD based on SMO+MPC controller is 8.87%. The latter has a smoother current waveform, smaller pulsation, and better stability of the motor.

[0229] Embodiment 3:

[0230] A deadbeat predictive current control device for a permanent magnet synchronous motor based on sliding mode anti-disturbance, comprising a memory and a processor, wherein the memory is used to store a computer program code and transmit the computer program code to the processor;

[0231] The processor is used to execute the permanent magnet synchronous motor zero-beat predictive current control method based on sliding mode anti-disturbance as described in Example 1 according to the instructions in the computer program code.

[0232] Embodiment 4:

[0233] A computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method for zero-beat predictive current control of a permanent magnet synchronous motor based on sliding mode anti-disturbance as described in Example 1.

[0234] Generally speaking, the computer instructions for implementing the method of the present invention may be carried in any combination of one or more computer-readable storage media. Non-transitory computer-readable storage media may include any computer-readable media, except for the signal itself that is temporarily propagating.

[0235] Computer-readable storage media, for example, may be, but are not limited to, electrical, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or devices, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection with one or more conductors, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with a system, device or device for executing instructions.

[0236] One or more programming languages ​​or their combinations can be used to write computer program codes for performing the operation of the present invention. These programming languages ​​include object-oriented programming languages, such as Java, Smalltalk, C++, and also include conventional procedural programming languages, such as C language or similar programming languages. In particular, Python language suitable for neural network calculations and platform frameworks based on TensorFlow, PyTorch, etc. can be used. The program code can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer and partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through various types of networks, including a local area network (LAN) or a wide area network (WAN), or an Internet connection is performed through an Internet service provider.

[0237] Although the embodiments of the present invention have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limitations on the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A deadbeat predictive current control method for a permanent magnet synchronous motor based on sliding mode anti-disturbance, characterized in that: The deadbeat prediction current control method is used to control an NPC structure three-level inverter circuit of a permanent magnet synchronous motor, and the deadbeat prediction current control method comprises the following steps: Step 1: Build a permanent magnet synchronous motor sliding mode anti-disturbance zero-beat predictive current vector control structure model: In the above formula: u d is the d-axis component of the stator voltage, u q is the q-axis component of the stator voltage, i d is the d-axis component of the stator current, i q is the q-axis component of the stator current, R s is the stator winding phase resistance, ω e is the rotor electrical angle, ψ d is the d-axis flux component, ψ q is the q-axis magnetic flux component, L s is the stator inductance, p=d / dt is the differential operator, ψ f is the permanent magnet flux, T e is the electromagnetic torque, n p is the pole pair number, T m is the load torque converted to the shaft end, J is the moment of inertia of the mechanical load converted to the shaft end, ω m is the rotor mechanical angular velocity, B is the viscous friction coefficient; Step 2: Establish the inner loop dq axis dual sliding mode anti-disturbance observer model based on the constant velocity reaching law, as follows: D-axis observer: Where: is the d-axis current component at the next moment, is the d-axis current component at the current moment, u q (k) is the q-axis current component at the current moment, is the estimated value of the total disturbance on the d-axis at the current moment, u d (k) is the d-axis voltage component at the current moment; Q-axis observer: Where: is the q-axis current component at the next moment, is the q-axis current component at the current moment, i d (k) is the d-axis current component at the current moment, u q (k) is the q-axis voltage component at the current moment, is the estimated value of the total disturbance on the q-axis at the current moment; Step 3: Establish a sliding mode controller model based on the model speed outer loop, and design the sliding mode controller using the exponential reaching law method: Where: is the differential of the sliding surface function, is the rate of change of the difference between the given speed and the feedback speed, is the second-order differential of the difference between the given speed and the feedback speed, C2 is an arbitrary positive real number, m is an arbitrary positive real number, ξ s >0, k>0; Step 4: Establish a current inner loop model based on the exponential reaching law to predict the current controller model. Predicted current equation: Where: e (k) is the electrical angular velocity of the rotor at the current moment; By shifting the terms in the above equation, we can solve for u d (k) and u q (k) value: At this time, the current inner loop model prediction current controller model based on the exponential reaching law is completed; Step 5: Construct an approximate two-level SVPWM deadbeat controller model based on virtual vector injection, use the output of the current controller model predicted by the current inner loop model in the exponential reaching law mode as the input of the approximate two-level SVPWM deadbeat controller model based on virtual vector injection, and use the improved SVPWM algorithm to reasonably allocate the duty cycle of each voltage vector to effectively reduce the switching loss of the inverter; Sector division: Divide the input voltage space vector into large and small sectors, and improve the SVPWM algorithm: Based on the principle of hiding the medium vector and large vector, and only using the zero vector, small vector and virtual vector for modulation, the area where the voltage space vector passes is divided into 12 asymmetric sectors; Formulate the switching sequence: Arrange the switching sequence of each small sector inverter to form a switching sequence chart. In order to reduce the stress of the switching devices, the 7-segment symmetrical voltage vector and the 9-segment symmetrical voltage vector are switched to each other. The variable sampling period is used to control the target space voltage vector V ref The sector number it falls into is used to formulate the action sequence table of the basic voltage vector; Vector action time calculation: determine the action time of each switching vector, form the SVPWM control model, and define the zero vector as V s0 , the small vector on the left side of the large sector is V s1 , the small vector on the right side of the large sector is V s2 When the 7-segment symmetrical voltage vector control is used, a two-level algorithm is used, using V s1 、V s2 、V s0 The target voltage vector V ref , V s1 、V s2 、V s0 The vector action time corresponds to T1, T2, and T0 respectively; when the 9-segment symmetrical voltage vector control is adopted, the approximate two-level algorithm of injecting virtual vector is adopted, and the virtual vector is further defined as V s3 , V s1 、V s2 Vector is used to synthesize virtual vector V s3 , use V s0 、V s1 、V s2 、V s3 The target voltage vector V ref , V s1 、V s2 、V s0 The vector action time corresponds to (T1′+T sw ′), T2′, T0′; T sw For a switching cycle, according to the vector synthesis law of different small sectors, the action time of each vector in each small sector is calculated respectively; then, according to the basic vector action time, the combined switch control pulse corresponding to each sector is obtained; Finally, according to the switching state and trigger time of each sector, the trigger time table of the switch devices in each sector is sorted out to form the SVPWM control model. The above control model is used to control the permanent magnet synchronous motor NPC inverter circuit.

2. The method for deadbeat predictive current control of a permanent magnet synchronous motor based on sliding mode anti-disturbance according to claim 1 is characterized in that: In the first step, the permanent magnet synchronous motor sliding mode anti-disturbance zero-beat predictive current control system uses the d-axis given current vector i d * = 0 rotor flux oriented vector control, the system consists of a speed outer loop and a current inner loop. The speed outer loop provides the stator current q-axis component given value i through a sliding mode controller. q * ,i d * 、i q * The observed values ​​of the d and q axis currents output by the sliding mode anti-disturbance observer Together they serve as the input of the deadbeat predictive current controller, and the output can obtain the control voltage V d With V q After the voltage is transformed into the αβ stationary coordinate system, the improved SVPWM technology is used to control the three-level NPC inverter to supply power to the permanent magnet synchronous motor; the current inner loop is based on the sliding mode anti-disturbance observer, and the real-time values ​​of the d and q axis currents measured by the current transformer are i d 、i q , and the rotor electrical angle ω e Together they serve as the input of the sliding mode anti-disturbance observer, and the output As the predictive current controller input, the rotor electrical angle ω e It is obtained by derivation of the rotor position angle θ.

3. The method for deadbeat predictive current control of a permanent magnet synchronous motor based on sliding mode anti-disturbance according to claim 2 is characterized in that: Step 2: Establish the inner loop dq axis dual sliding mode anti-disturbance observer model based on the constant velocity reaching law, assuming that the disturbance caused by the permanent magnet synchronous motor parameters is bounded and in a very short sampling period T sw The voltage inside is constant, and the voltage equations of the motor's d-axis and q-axis can be expressed as: Where: f d is the disturbance caused by the d-axis parameter, f q is the disturbance caused by the q-axis parameter, F d f d The rate of change, F q f q The rate of change, △Ls is the stator inductance disturbance, △Rs is the stator resistance disturbance, Δψ f is the stator flux disturbance; i d and f d is the state variable, ω e is the rotor electrical angular velocity; Taking the d-axis feedback current as the observation object, a sliding mode anti-disturbance observer is designed: Where: is the rate of change of the estimated d-axis current, is the rate of change of the d-axis disturbance estimate, is the d-axis current, is the estimated value of the d-axis disturbance, m is the sliding mode control gain (m>0), e id is the error between the d-axis observed current and the feedback current u(e id ) represents the sliding mode control rate of the d-axis; Compensating the d-axis disturbance value into the voltage equation yields: Discretize the above formula and finally get the d-axis observer: Where: is the d-axis current component at the next moment, is the d-axis current component at the current moment, i q (k) is the q-axis current component at the current moment, is the estimated value of the total disturbance on the d-axis at the current moment, u d (k) is the d-axis voltage component at the current moment; Similarly, taking the q-axis feedback current as the observation object, the sliding mode anti-disturbance observer is designed as follows: Where: is the rate of change of the estimated value of the q-axis current, is the rate of change of the estimated value of the q-axis disturbance, is the q-axis current, is the estimated value of the q-axis disturbance, n is the sliding mode control gain (n>0), e iq is the error between the q-axis observed current and the feedback current u(e iq ) is the q-axis sliding mode control rate; Compensating the d-axis disturbance value into the voltage equation yields: Discretize the above formula to get the q-axis observer: Where: is the q-axis current component at the next moment, is the q-axis current component at the current moment, i d (k) is the d-axis current component at the current moment, u q (k) is the q-axis voltage component at the current moment, is the estimated value of the total disturbance on the q-axis at the current moment; At this point, the inner loop dq axis dual sliding mode anti-disturbance observer model based on the constant velocity reaching law is completed.

4. The method for deadbeat predictive current control of a permanent magnet synchronous motor based on sliding mode anti-disturbance according to claim 3 is characterized in that: Step 3: Establish a model of speed outer loop sliding mode controller based on the model. To speed up the sliding surface speed of the system motion point and reduce the chattering near the sliding surface, the speed outer loop sliding mode controller is designed based on the exponential approach law; define the controller state variables: Where: m * For a given rotor mechanical angular velocity, ω m The actual rotor mechanical angular velocity measured by the encoder is the feedback rotor mechanical angular velocity, e ω is the difference between the given rotor mechanical angular velocity and the feedback rotor mechanical angular velocity, is the rate of change of the difference between the given speed and the feedback speed, is the rate of change of a given speed; The sliding mode controller is designed using the exponential reaching law: Where: is the differential of the sliding surface function, is the second-order differential of the difference between the given speed and the feedback speed, C2 is an arbitrary positive real number, m is an arbitrary positive real number, ξ s >0, k>0; At this point, the model construction of the model speed outer loop sliding mode controller is completed.

5. The method for deadbeat predictive current control of a permanent magnet synchronous motor based on sliding mode anti-disturbance according to claim 4 is characterized in that: Step 4: Establish a current inner loop model based on the exponential reaching law to predict the current controller model. The current inner loop model based on the exponential reaching law is established to predict the current controller model, and the voltage equation of the permanent magnet synchronous motor in the synchronous rotating coordinate system is: In a very small sampling period T sw In the above formula, the d-axis and q-axis currents are discretized to obtain: Use discrete equations to predict the current i at the next moment d (k+1) and i q (k+1), assuming that the d-axis current given value is i d * =i d (k+1), q-axis current given value i q * =i q (k+1); with d and q axis stator voltage given value u d 、u q is the control rate, and the predicted current equation is obtained: Where: e (k) is the electrical angular velocity of the rotor at the current moment; By shifting the terms in the above equation, we can solve for u d (k),u q (k) value: At this point, the current inner loop model predictive current controller model based on the exponential reaching law is completed.

6. A deadbeat predictive current control system for a permanent magnet synchronous motor based on sliding mode anti-disturbance, characterized in that: The deadbeat predictive current control system of the permanent magnet synchronous motor based on sliding mode anti-disturbance is a deadbeat predictive current control model constructed by the method described in any one of claims 1 to 5; The control system comprises: Deadbeat predictive current vector controller: a mathematical model for the synchronous motor sliding mode anti-disturbance deadbeat predictive current vector control; Inner loop dq axis dual sliding mode anti-disturbance observer: Algorithm for executing the inner loop dq axis dual sliding mode anti-disturbance observer; Outer loop sliding mode controller: used to execute the model speed outer loop sliding mode control algorithm; Predictive current controller: used to execute the current inner loop model algorithm based on the exponential reaching law method; Approximate two-level SVPWM deadbeat controller based on injected virtual vector: used to implement the improved SVPWM algorithm.

7. A permanent magnet synchronous motor deadbeat predictive current control device based on sliding mode anti-disturbance, characterized in that: The method comprises a memory and a processor, wherein the memory is used to store computer program code and transmit the computer program code to the processor; The processor is configured to execute the method according to any one of claims 1 to 5 according to instructions in the computer program code.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in any one of claims 1 to 5 is implemented.