Permanent magnet synchronous motor low-buffeting terminal sliding mode control method based on adaptive disturbance compensation

Abstract

The invention relates to the technical field of permanent magnet synchronous motors, and particularly discloses a permanent magnet synchronous motor low-buffeting terminal sliding mode control method based on adaptive disturbance compensation, comprising the following steps: S01, redefining a motor rotational inertia mathematical model; s02, improving a traditional nonsingular fast terminal sliding mode surface based on nonlinear system representation, and constructing a novel dynamic gain nonsingular fast terminal sliding mode surface; s03, performing derivation on the novel dynamic gain nonsingular fast terminal sliding mode surface, and combining a derivation function with nonlinear system representation to obtain an equivalent control law ieq; and S04, adopting an exponential reaching law isw as a switching control law of the sliding mode controller, and obtaining a novel different fast terminal sliding mode controller model according to an equivalent control principle in combination with the equivalent control law ieq and the exponential reaching law isw. The invention aims to improve the tracking response speed of a position instruction during servo control of the permanent magnet synchronous motor.

Description

Technical Field

[0001] The present application relates to the technical field of permanent magnet synchronous motors, and specifically discloses a low-jitter terminal sliding mode control method for permanent magnet synchronous motors based on adaptive disturbance compensation. Background Art

[0002] Permanent Magnet Synchronous Machine (PMSM) is a motor that works on the principle of electromagnetic induction.

[0003] A permanent magnet synchronous motor consists of a stator, rotor, and end caps. Its use of permanent magnets for excitation simplifies the motor's structure, reduces processing and assembly costs, and eliminates slip rings and brushes, which are prone to problems, improving motor reliability. Furthermore, since no excitation current is required, there is no excitation loss, which improves the motor's efficiency and power density.

[0004] Permanent magnet synchronous motors are widely used in industrial robots, CNC machine tools, and other fields. High-precision servo control systems based on permanent magnet synchronous motors are nonlinear and strongly coupled systems, and the control accuracy of traditional proportional-integral-differential controllers is generally limited. Sliding mode control has been widely used due to its advantages such as simple structure, fast dynamic response, and strong robustness.

[0005] Sliding mode control is divided into the approaching phase and the sliding phase, and the approaching law controls the approaching speed. Traditional sliding mode control has discontinuous switching characteristics, which will cause system chattering. Essentially, this is caused by the non-ideal approach of the system trajectory to the sliding surface. In order to reduce the chattering problem in the control process, the approaching law is usually improved to improve the dynamic quality of the system's approaching motion. Traditional approaching laws often require a compromise between the degree of chattering and the response speed. Especially when the time-varying disturbance is large, suppressing the disturbance by increasing the switching gain at the expense of the response speed will inevitably aggravate the chattering. Improper parameter selection may even lead to system instability. Therefore, the use of disturbance observer compensation can solve the problem of unknown disturbances on the one hand, and reduce the switching gain on the other hand, thereby suppressing chattering. The convergence characteristics of the system state in the sliding phase are determined by the sliding surface and are related to the response speed of the system. Compared with the infinite time asymptotic convergence characteristics of the traditional linear sliding surface, the terminal sliding surface enables the system state to converge in a finite time and has a fast convergence speed. It can meet the requirements of the fast convergence characteristics of the state of high-precision motor control systems and has good research and application prospects.

[0006] Traditional non-singular fast terminal sliding surfaces use fixed convergence gains. When the system moves away from and approaches the sliding surface, the nonlinear and linear terms dominate the rapid convergence, respectively, making it impossible to dynamically adjust adaptively to the system state or disturbance state. Furthermore, when the control system is subject to external disturbances or parameter perturbations, if no compensation measures are taken for the disturbance estimation, the sliding mode controller must have a large switching gain. Excessive switching gain can lead to severe chattering, seriously impairing the system's control performance. Furthermore, some current solutions use state variable-based disturbance observers to observe and compensate for disturbances. These methods often rely on precise mathematical models of the controlled object, significantly reducing estimation accuracy in the presence of unmodeled dynamics or parameter perturbations, and exhibiting poor adaptability to dynamic disturbances. This is particularly true for nonlinear external disturbances, where observation performance is poor, leading to reduced control performance.

[0007] In view of this, the present invention provides a low-jitter terminal sliding mode control method for a permanent magnet synchronous motor based on adaptive disturbance compensation to solve the above problems. Summary of the Invention

[0008] The purpose of this invention is to provide a control algorithm that suppresses chattering in sliding-mode control of permanent magnet synchronous motors. This algorithm improves the sliding surface in conventional sliding-mode control systems and enhances the tracking response speed of position commands during servo control of permanent magnet synchronous motors. By adaptively compensating for interference, the algorithm reduces the switching gain of the sliding-mode control, effectively suppressing chattering during conventional sliding-mode control and reducing speed fluctuations in the sliding-mode control system, thereby improving the interference resistance and robustness of the motor servo control system.

[0009] In order to achieve the above object, the present invention provides the following basic scheme:

[0010] A low-jitter terminal sliding mode control method for a permanent magnet synchronous motor based on adaptive disturbance compensation comprises the following steps:

[0011] S01: According to the mechanical equation of the permanent magnet synchronous motor and the dq-axis electromagnetic torque equation, the mathematical model of the motor moment of inertia is obtained, and the system state variable x1 is constructed as the position reference instruction θ * and the tracking error of the actual value θ, redefine the mathematical model of the motor's moment of inertia and obtain its nonlinear system representation;

[0012] S02: Based on the above nonlinear system representation, the traditional nonsingular fast terminal sliding surface is improved to construct a new dynamic gain nonsingular fast terminal sliding surface;

[0013] S03: Derivative the new dynamic gain non-singular fast terminal sliding mode surface to obtain the derivative function, and express the derivative function with the nonlinear system in step S01 to obtain the equivalent control law i eq ;

[0014] S04: Using exponential reaching law i swAs a switching control law of the modified sliding mode controller, a new type of nonsingular fast terminal sliding mode controller model is obtained according to the equivalent control principle combined with the equivalent control law i eq and the exponential reaching law i sw .

[0015] Further, the mathematical model of the motor moment of inertia is Wherein, J is the motor moment of inertia; B is the friction coefficient; P n is the number of pole pairs; ψ f is the permanent magnet flux linkage; ω is the angular velocity; i q is the q-axis current; T L is the load torque; d is the disturbance, x1=θ * -θ; the system state variable x2 is the derivative of the tracking error, After redefining the mathematical model of the motor moment of inertia, the nonlinear system is represented as:

[0016]

[0017] Further, the new type of dynamic gain nonsingular fast terminal sliding surface is:

[0018]

[0019] Wherein, the coefficients α, β, χ, λ are positive numbers, m1, m2>1, p, q are positive odd numbers, and satisfy 1

[0020] Further, when s=0, the state variable x2 is: When the system state deviates from the equilibrium point, the convergence speed of the error is determined by ; when the system state is close to the sliding surface and converges to the equilibrium point, the convergence speed is determined by -tanh(λx1), and the dynamic gain function related to the system state is introduced to accelerate the convergence speed when the error is large.

[0021] Further, in step S03, the derivative function is:

[0022] Wherein,

[0023]

[0024] Further, assuming that the real disturbance d of the system can be estimated, the estimated value is Let According to the derivative function in step S03 and the nonlinear system representation, the equivalent control law i eq is obtained as: According to the exponential reaching law i sw As the switching control law of the sliding mode controller, we can obtain In the formula, coefficient k, ε>0, combined with the equivalent control law i eq and exponential reaching law i sw The new non-singular fast terminal sliding mode controller model is obtained as

[0025] Furthermore, in view of the strong nonlinearity of the disturbance value d in practice, a radial basis neural network is used to approximate the unknown disturbance d of the system. The state variable x = [x1, x2] is used as the input of the radial basis neural network. The radial basis neural network hidden layer is designed to have n nodes, and the output of its Gaussian function is expressed as h = [h1 h2…h n ], and the jth node is represented as c j is the center vector of the hidden neuron, b j is the width of the Gaussian basis function, and is the estimated value of the ideal weight matrix W, and is defined is the optimal weight estimation error, then the estimated value of disturbance d is The estimated error is:

[0026]

[0027] Where η is the approximation error of the radial basis neural network, |η|≤η N ;

[0028] Design the adaptive law of radial basis neural network weights as:

[0029]

[0030] Furthermore, the new non-singular fast terminal sliding mode controller model is To prove the stability, construct the Lyapunov function as Combined with the following formula

[0031]

[0032]

[0033] Substitute it into the differential equation and simplify it:

[0034]

[0035] Furthermore, the new non-singular fast terminal sliding mode controller model is designed Convergence time is finite.

[0036] Combined with the following formula:

[0037]

[0038] Construct Lyapunov function V = 0.5s 2 , and taking its derivative we can get:

[0039]

[0040] The principle and effect of this solution are:

[0041] 1. Compared with existing technologies, this method can effectively improve equipment performance and expand its application potential by improving the dynamic response and steady-state accuracy of the permanent magnet synchronous motor control system and reducing sliding mode control chattering.

[0042] 2. Compared with the existing technology, this method constructs a new dynamic gain non-singular fast terminal sliding surface and introduces a dynamic gain related to the error variable. It accelerates convergence in the early stage of convergence and reduces chattering when approaching the equilibrium point by avoiding excessive convergence gain. It can improve the convergence performance of the traditional non-singular fast terminal sliding surface.

[0043] 3. Compared with existing technologies, this approach improves the convergence characteristics of the position tracking error by designing a dynamic gain nonsingular fast terminal sliding surface related to the position tracking error, dynamically adjusting the convergence rate of the position tracking error. Based on this novel sliding surface, a new sliding mode controller is designed, which performs online adaptive disturbance compensation. This adaptive disturbance compensation control system effectively reduces chattering, improves speed accuracy, and enhances the dynamic response of position tracking.

[0044] 4. Compared with the existing technology, the control method in which the disturbance is adaptively compensated by the radial basis function neural network in this method can accurately estimate the nonlinear disturbance and effectively reduce the speed fluctuation during the operation of the permanent magnet synchronous motor. Even when the switching gain is greatly reduced, the position command information can still be tracked with high precision to reduce the vibration. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0046] Figure 1 A block diagram of a low-jitter terminal sliding mode control system of a permanent magnet synchronous motor low-jitter terminal sliding mode control method based on adaptive disturbance compensation proposed in an embodiment of the present application is shown;

[0047] Figure 2A diagram showing the effect of disturbance values ​​in a low-jitter terminal sliding mode control method for a permanent magnet synchronous motor based on adaptive disturbance compensation proposed in an embodiment of the present application is shown;

[0048] Figure 3 A schematic diagram showing the speed and error effects in a low-jitter terminal sliding mode control method for a permanent magnet synchronous motor based on adaptive disturbance compensation proposed in an embodiment of the present application is shown;

[0049] Figure 4 A schematic diagram of position and error effects in a low-jitter terminal sliding mode control method for a permanent magnet synchronous motor based on adaptive disturbance compensation proposed in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0050] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0051] Implementation example Figures 1-4 As shown:

[0052] A low-jitter terminal sliding mode control method for a permanent magnet synchronous motor based on adaptive disturbance compensation comprises the following steps:

[0053] S01: According to the permanent magnet synchronous motor mechanical equation and dq axis electromagnetic torque equation,

[0054] as follows: Obtain the mathematical model of the motor's moment of inertia and construct the system state variable x1 as the position reference instruction θ * and the tracking error of the actual value θ, redefine the mathematical model of the motor's moment of inertia and obtain its nonlinear system representation;

[0055] first:

[0056] According to the mechanical equation of permanent magnet synchronous motor and the dq-axis electromagnetic torque equation, its mathematical model is:

[0057] Where, J is the motor's moment of inertia; B is the friction coefficient; P n is the pole pair number; ψ f is the permanent magnet flux; ω is the angular velocity; i q is the q-axis current; T L is the load torque; d is the disturbance.

[0058] Construct the system state variable x1 as the position reference instruction θ * and the tracking error of the actual value θ, x1 = θ * -θ; the system state variable x2 is the derivative of the tracking error,

[0059] The nonlinear system can be reformulated as:

[0060]

[0061] S02: Based on the above nonlinear system representation, the traditional nonsingular fast terminal sliding surface is improved to construct a new dynamic gain nonsingular fast terminal sliding surface;

[0062] Specifically:

[0063] In order to obtain fast convergence performance in finite time, the traditional non-singular fast terminal sliding surface is improved, and a new dynamic gain non-singular fast terminal sliding surface is constructed as follows:

[0064]

[0065] Wherein, coefficients α, β, χ, and λ are positive numbers, m1 and m2 are greater than 1, and p and q are positive odd numbers, satisfying 1<p / q<2.

[0066] When s=0, the state variable x2 is:

[0067]

[0068] according to When the system state is far away from the equilibrium point, the position error is large, and the error convergence speed is mainly determined by When the system state is close to the sliding surface and converges to the equilibrium point, in order to prevent the error caused by the power term from converging too slowly, the convergence speed is determined by -tanh(λx1). At the same time, a dynamic gain function related to the system state is introduced. It can speed up the convergence when the error is large and avoid chattering caused by excessive convergence gain when the error is small.

[0069] S03: Derivative the new dynamic gain non-singular fast terminal sliding mode surface to obtain the derivative function, and express the derivative function with the nonlinear system in step S01 to obtain the equivalent control law i eq ;

[0070] Specifically:

[0071] Pair Derivative of sliding surface function:

[0072]

[0073] Where,

[0074]

[0075] According to the definition of each coefficient, it is obvious that the derivative of the sliding surface is nonsingular, and the sliding surface satisfies the nonsingular fast terminal sliding surface. First, assume that the real disturbance d of the system can be estimated, and the estimated value is Let

[0076] According to

[0077]

[0078] The equivalent control law i eq is:

[0079]

[0080] S04: Adopt the exponential reaching law i sw As the switching control law of the modified sliding mode controller, according to the equivalent control principle combined with the equivalent control law i eq and the exponential reaching law i sw a new type of nonsingular fast terminal sliding mode controller model is obtained.

[0081] Specifically:

[0082] Adopt the exponential reaching law as the switching control law of the modified sliding mode controller, and the expression is:

[0083]

[0084] In the formula, the coefficients k and ε are greater than 0.

[0085] According to the equivalent control principle, according to the formula The formula The new type of nonsingular fast terminal sliding mode controller model is designed as:

[0086]

[0087] The equivalent control law can ensure the tracking performance of the system, the switching control law handles the residual error, the disturbance compensation actively offsets the estimated disturbance, which can greatly reduce the switching gain and reduce the chattering caused by switching control and nonlinear disturbance. The above sliding mode controller model assumes that the disturbance value is effectively estimated. Considering the strong nonlinearity of the disturbance value in practice, since the neural network can approximate any continuous function of a nonlinear system with arbitrary precision, in order to facilitate the implementation of the scheme, a radial basis neural network is used to approximate the unknown disturbance d of the system.

[0088] Specifically:

[0089] It is known that the radial basis neural network is a neural network with a single hidden layer. The state variable x = x1, x2] is used as the input of the radial basis neural network. Assuming that the designed radial basis neural network hidden layer has n nodes, the output of its Gaussian function is expressed as h = [h1 h2…h n ], and the jth node is represented as c j is the center vector of the hidden neuron, b j is the width of the Gaussian basis function.

[0090] is the estimated value of the ideal weight matrix W, and is defined is the optimal weight estimation error, then the estimated value of disturbance d is The estimated error is:

[0091]

[0092] Where η is the approximation error of the radial basis neural network, |η|≤η N .

[0093] The update of the weight matrix in the radial basis neural network plays an important role in the tracking accuracy of the disturbance d. The weight adaptation law of the radial basis neural network is designed as:

[0094]

[0095] Finally, we need to verify the stability and finite convergence time:

[0096] Regarding stability:

[0097] The model of the new non-singular fast terminal sliding mode controller is To prove the stability, construct the Lyapunov function as Combined with the following formula

[0098]

[0099] Substitute it into the differential equation and simplify it:

[0100]

[0101] According to the constraint condition, f(x1),k>0, select the parameter η N ≥|ε|, According to Lyapunov's second stability theory, the tracking error of the system converges asymptotically and the designed controller model is stable.

[0102] Regarding the finite convergence time:

[0103] A new non-singular fast terminal sliding mode controller model is designed Convergence time is finite.

[0104] Combined with the following formula:

[0105]

[0106] Construct Lyapunov function V = 0.5s 2 , and taking its derivative we can get:

[0107]

[0108] Specifically:

[0109] pass It can be seen that there exists a positive constant ζ that satisfies

[0110] Derivation to obtain By solving the problem, we can know that the convergence time t of the designed new non-singular terminal sliding surface is c satisfy The proof of finite-time convergence of the system is completed.

[0111] In summary, Figure 1 The block diagram of the low-jitter terminal sliding mode control system of permanent magnet synchronous motor based on adaptive disturbance compensation is provided. The servo control system adopts the composite control form of "outer loop sliding mode controller + inner loop current PI controller". The outer loop sliding mode controller collects position command information θ * With the actual position information of the motor θ, the tracking error x1 and its first-order derivative x2 of the position information are solved, and the

[0112] The reaching law and the equivalent control law are calculated as the output of the sliding mode controller. Among them, the radial basis neural network output As:

[0113]

[0114] The estimate of the disturbance value in Used to adaptively update the output weight estimation derivative of the radial basis neural network This allows for rapid compensation of disturbances, affecting the output of the equivalent control law. The output of the outer sliding mode controller serves as the input to the inner q-axis current loop, while the d-axis current loop is set to zero. The output of the current loop undergoes a dq-αβ coordinate transformation, converting it into dq-axis currents. This is then converted into six PWM pulses using the SVPWM algorithm to control the switching of the power modules in the inverter bridge.

[0115] The advantage of the present invention is that it improves the convergence characteristics of the position tracking error by designing a dynamic gain non-singular fast terminal sliding surface related to the position tracking error, and dynamically adjusts the convergence speed of the position tracking error. A new sliding mode controller is designed based on the designed new sliding surface and the disturbance is adaptively compensated online. The control system after adaptive disturbance compensation can effectively reduce jitter and improve speed accuracy, and improve the dynamic response effect of position tracking. In order to verify the effectiveness of the present invention, the scheme has been verified. In a permanent magnet synchronous motor servo control system with nonlinear disturbances related to the speed, the disturbance estimation, position tracking and speed fluctuation effects of the system with disturbance compensation and uncompensated disturbance are compared. Figures 2-4 , the switching gain ε in the uncompensated scheme is 1000, and the switching gain after compensation is set to 200. The results show that the control method based on the present invention in which the disturbance is adaptively compensated by the radial basis function neural network can accurately estimate nonlinear disturbances and effectively reduce the speed fluctuations during the operation of the permanent magnet synchronous motor. Even when the switching gain is greatly reduced, the position command information can still be tracked with high precision to reduce jitter. In summary, the present invention has the significant advantages of low jitter and fast dynamic response in the servo control system driven by the permanent magnet synchronous motor using sliding mode control.

[0116] The key points of this method:

[0117] 1. A new dynamic gain non-singular fast terminal sliding surface is constructed. The dynamic gain related to the error variable is introduced to accelerate convergence in the early stage of convergence. When approaching the equilibrium point, excessive convergence gain is avoided and chattering is reduced. This can improve the convergence performance of the traditional non-singular fast terminal sliding surface.

[0118] 2. Through position instruction information θ * The actual position information θ of the motor is collected, and an adaptive sliding mode controller based on a novel dynamic gain non-singular fast terminal sliding surface is constructed. The disturbances applied to the permanent magnet synchronous motor are approximated and compensated in the sliding mode controller through an adaptive radial basis function neural network. The weight matrix of the radial basis function neural network is updated online through the adaptive law.

[0119] 3. The initial weight matrix of the adaptive radial basis neural network can be set to a zero matrix, and the initial value c of the center vector of its hidden layer neurons can be selected by the actual value of the disturbance collected and the K-means clustering method. j and the initial value b of the Gaussian basis function width j .

[0120] 4. The stability of the designed new sliding mode controller is proved by Lyapunov's theorem, and it is proved that the system state variables can converge in finite time under the improved dynamic gain non-singular fast terminal sliding mode.

[0121] The application aims to provide a control algorithm for inhibiting chattering of a sliding mode control of a permanent magnet synchronous motor, improving a sliding mode surface in a conventional sliding mode control system, and improving a tracking response speed of a position instruction in a servo control of the permanent magnet synchronous motor.

[0122] The above is only a preferred embodiment of the present application, and does not limit the present application in any form. Although the present application has been disclosed as above with the preferred embodiment, it is not intended to limit the present application. Any person skilled in the art can make some changes or modifications to the above disclosed technical content to obtain equivalent embodiments with equivalent changes, without departing from the technical solution of the present application. Any simple modification, equivalent change and modification of the above embodiments, which are made according to the technical essence of the present application, still belong to the scope of the technical solution of the present application.

Claims

1. A low-jitter terminal sliding mode control method for a permanent magnet synchronous motor based on adaptive disturbance compensation, characterized in that: It includes the following steps: S01: According to the mechanical equation of the permanent magnet synchronous motor and the dq-axis electromagnetic torque equation, the mathematical model of the motor moment of inertia is obtained, and the system state variable x1 is constructed as the position reference instruction θ * and the tracking error of the actual value θ, redefine the mathematical model of the motor's moment of inertia and obtain its nonlinear system representation; S02: Based on the above-mentioned representation of the nonlinear system, improve the traditional non-singular fast terminal sliding surface to construct a new type of dynamic gain non-singular fast terminal sliding surface; S03: Derivative the new dynamic gain non-singular fast terminal sliding mode surface to obtain the derivative function, and combine the derivative function with the nonlinear system representation in step S01 to obtain the equivalent control law i eq ; S04: Using exponential reaching law i sw As the switching control law of the sliding mode controller, according to the equivalent control principle combined with the equivalent control law i eq and exponential reaching law i sw A new non-singular fast terminal sliding mode controller model is obtained.

2. The method for low-jitter terminal sliding mode control of a permanent magnet synchronous motor based on adaptive disturbance compensation according to claim 1, characterized in that: The mathematical model of the motor's moment of inertia is: Where: J is the motor's moment of inertia; B is the friction coefficient; P n is the pole pair number; ψ f is the permanent magnet flux; ω is the angular velocity; i q is the q-axis current; T L is the load torque; d is the disturbance, and the constructed system state variable x1 is the position reference instruction θ * and the tracking error of the actual value θ, x1 = θ * -θ; the system state variable x2 is the derivative of the tracking error, After redefining the mathematical model of motor moment of inertia, its nonlinear system is expressed as:

3. The low-jitter terminal sliding mode control method for a permanent magnet synchronous motor based on adaptive disturbance compensation according to claim 2, characterized in that: The functional expression of the new type of dynamic gain non-singular fast terminal sliding surface is: In the formula, the coefficients α, β, χ, λ are positive numbers respectively, m1, m2 > 1, p, q are positive odd numbers, and satisfy 1 < p / q < 2, and s is the new type of dynamic gain non-singular fast terminal sliding surface.

4. The method for low-jitter terminal sliding mode control of a permanent magnet synchronous motor based on adaptive disturbance compensation according to claim 3 is characterized in that: When s=0, the state variable x2 is: When the system state is far away from the equilibrium point, the convergence rate of the error is determined by Determine; when the system state is close to the sliding surface and converges to the equilibrium point, the convergence speed is determined by -tanh(λx1), and the dynamic gain function related to the system state is introduced Speed ​​up the convergence when the error is large.

5. The method for low-jitter terminal sliding mode control of a permanent magnet synchronous motor based on adaptive disturbance compensation according to claim 4, characterized in that: In step S03, the derivative function is: in, 6. The method for low-jitter terminal sliding mode control of a permanent magnet synchronous motor based on adaptive disturbance compensation according to claim 5, characterized in that: Assume that the real disturbance d of the system can be estimated, and the estimated value is make According to the derivative function in step S03 and the nonlinear system representation, the equivalent control law i can be obtained. eq for: According to the exponential reaching law i sw As the switching control law of the sliding mode controller, we can obtain In the formula, coefficient k, ε>0, combined with the equivalent control law i eq and exponential reaching law i sw The new non-singular fast terminal sliding mode controller model is obtained as 7. The method for low-jitter terminal sliding mode control of a permanent magnet synchronous motor based on adaptive disturbance compensation according to claim 6, characterized in that: In view of the strong nonlinearity of the disturbance value d, a radial basis neural network is used to approximate the unknown disturbance d of the system. The state variable x = [x1, x2] is used as the input of the radial basis neural network. The radial basis neural network hidden layer is designed to have n nodes, and the output of its Gaussian function is expressed as h = [h1, h2, ..., h n ], and the jth node is represented as c j is the center vector of the hidden neuron, b j is the width of the Gaussian basis function, and is the estimated value of the ideal weight matrix W, and is defined is the optimal weight estimation error, then the estimated value of disturbance d is The estimated error is: Where η is the approximation error of the radial basis neural network, |η|≤η N ; Design the weight adaptive law of the radial basis neural network as:

8. The method for low-jitter terminal sliding mode control of a permanent magnet synchronous motor based on adaptive disturbance compensation according to claim 7, characterized in that: The model of the new non-singular fast terminal sliding mode controller is To prove the stability, construct the Lyapunov function as Combine the following formula: After substituting it into its differential equation and simplifying:

9. The method for low-jitter terminal sliding mode control of a permanent magnet synchronous motor based on adaptive disturbance compensation according to claim 8, characterized in that: A new non-singular fast terminal sliding mode controller model is designed Convergence time is limited to prove: Combine the following formula: Construct Lyapunov function V = 0.5s 2 , and taking its derivative we can get:

Images