Permanent magnet synchronous motor data-driven adaptive iterative learning control method based on event trigger mechanism

Abstract

The invention discloses a permanent magnet synchronous motor data-driven adaptive iterative learning control method based on an event trigger mechanism. The method comprises the following four steps: constructing a kinetic model of a permanent magnet synchronous motor system, determining a motor discrete state space model, designing a data-driven adaptive iterative learning controller based on an optimized idea, and setting the event trigger mechanism. According to the method, under the condition that parameters of the permanent magnet synchronous motor system are uncertain, on the premise that it is guaranteed that tracking errors of the permanent magnet synchronous motor system converge to be close to zero within limited iteration times, the updating times of the controller are reduced, and the requirement for system hardware performance is lowered; and meanwhile, the controller is designed only by using input and output measurement data generated by the dynamical model without depending on specific parameters of the dynamical model, so that the difficulty in model decoupling is reduced, and practical engineering popularization is facilitated.

Description

Technical Field

[0001] The invention belongs to the technical field of motor optimization control and proposes a data-driven adaptive iterative learning control method for a permanent magnet synchronous motor based on an event trigger mechanism. Background Art

[0002] Permanent magnet synchronous motors have the advantages of simple structure, small size, light weight, high efficiency, precise positioning, and easy control. They are widely used in industrial automation, electric vehicles, household appliances and other fields, and play an irreplaceable role in situations that require high efficiency, high power density and precise control.

[0003] For permanent magnet synchronous motors that perform repetitive tasks, iterative learning control is a control strategy that can effectively improve trajectory tracking accuracy. However, most of the existing research results are a series of control methods proposed for permanent magnet synchronous motors during batch operation, which require a large memory to store the data information generated during the operation of the permanent magnet synchronous motor. However, due to limited communication bandwidth and reliability considerations, it is necessary to reduce the amount of data transmission. Event triggering, as a common technical means to solve this problem, has great advantages in practical applications. It can not only meet the requirements of system tracking performance, but also reduce the transmission burden. Therefore, for permanent magnet synchronous motors, there is a problem of designing an iterative learning controller based on event triggering.

[0004] In actual production applications, the performance of permanent magnet synchronous motors depends largely on the accuracy of their parameter models. However, permanent magnet synchronous motors are affected by various external uncertainties, making it difficult for operators to accurately grasp the model information of permanent magnet synchronous motors. Especially in the production process of high-precision products, accurate parameter models are the key basis for permanent magnet synchronous motors to achieve high-performance control. Inaccurate parameter models will lead to difficulties in optimizing the control strategy of permanent magnet synchronous motors, thereby affecting key performance indicators such as output power, torque response, and efficiency of permanent magnet synchronous motors. When the system model of permanent magnet synchronous motors is uncertain or even unknown, the data-driven adaptive iterative learning control method can enable the speed of permanent magnet synchronous motors to achieve high tracking accuracy within a limited time interval. In view of the fact that the model information of permanent magnet synchronous motors is completely unknown or partially unknown, there are few relevant research results that can well deal with the influence of unknown model information on the control performance of permanent magnet synchronous motors, achieve good control performance of permanent magnet synchronous motors, and reduce the storage of data information. Therefore, it is a very meaningful study to design a data-driven adaptive iterative learning control method without relying on the accurate information of the permanent magnet synchronous motor system model, so that the speed of the permanent magnet synchronous motor can quickly track the expected speed and effectively save control resources. Summary of the invention

[0005] In order to overcome the shortcomings of the prior art, to reduce the dependence of the permanent magnet synchronous motor on the physical model information in the actual application, and to enable the permanent magnet synchronous motor to quickly reach a stable state and have good robustness, the present invention proposes a data-driven adaptive iterative learning control method for a permanent magnet synchronous motor based on an event trigger mechanism, which uses the optimization objective function technology to estimate the unknown parameter items in the permanent magnet synchronous motor system, and then designs a parameter adaptive update algorithm, and designs an adaptive iterative learning controller based on the estimated parameter information and the input and output data of the system; in addition, an event trigger mechanism is designed to reduce the requirements of the permanent magnet synchronous motor controller on the system hardware performance. This design can effectively reduce the dependence of the permanent magnet synchronous motor control method on the precise model information, while ensuring that the permanent magnet synchronous motor system has a fast convergence capability and reducing the storage of data information.

[0006] The technical solutions proposed to solve the above technical problems are as follows:

[0007] A data-driven adaptive iterative learning control method for a permanent magnet synchronous motor based on an event trigger mechanism comprises the following steps:

[0008] Step 1: Establish the dynamic model of permanent magnet synchronous motor:

[0009]

[0010] Wherein, t represents the time point, p(t) represents the rotor angular position of the permanent magnet synchronous motor, and w(t) represents the angular velocity of the permanent magnet synchronous motor during movement; and represent the first-order derivatives of p(t) and w(t), respectively; B is the viscous friction coefficient, J is the moment of inertia, i q is the q-axis stator current, n p is the number of pole pairs of the permanent magnet synchronous motor, ψ f Represents magnetic flux, T L represents the load torque; in order to facilitate the design of the subsequent controller, let u(t) = i q , represents the input value of the system;

[0011] Step 2: Determine the discrete state space model of the permanent magnet synchronous motor system. The process is as follows:

[0012] 2.1 By selecting the sampling period T s , and use the Euler method to calculate the continuous system model (1) and Discretize it and get:

[0013]

[0014] 2.2 Substituting equation (2) into equation (1), after mathematical transformation, the discrete system model expression of the permanent magnet synchronous motor in motion is obtained as follows:

[0015]

[0016] Where v∈T N is a discrete time point; T N ={0,1,…,N}, N is a finite positive integer;

[0017] 2.3 Define the position p(v) and speed w(v) of the permanent magnet synchronous motor as the state variable x(v), expressed as: x(v) = [p(v)w(v)] Τ , the input variable of the permanent magnet synchronous motor system is defined as the current u(v), and the output variable is the angular velocity w(v); considering that the above permanent magnet synchronous motor system is repeatedly operated in a finite time, the permanent magnet synchronous motor model described by equation (3) is expressed as the following discrete state space model:

[0018]

[0019] Where k∈N + Indicates the number of system iterations, u k (v)∈R and y k (v)∈R represent the bounded single input and single output vectors of the system respectively; x k (v)∈R 2 Represents the state vector of the system; A, B, C and D are parameter matrices with appropriate dimensions, which contain various model information of the permanent magnet synchronous motor and satisfy CB≠0, in the following form:

[0020]

[0021] In order to reduce the dependence on physical model information in the actual application of permanent magnet synchronous motor, the situation where the information of parameter matrices A, B, C and D is unknown or incompletely known is considered. Only the input and output data information in equation (4) is used to design the controller to realize data-driven control.

[0022] 2.4 To facilitate the design of subsequent controllers, input u k (v), output y k (v) Expressed in supervector form: y k =[y k (1),y k (2),…,y k (N)] T ,u k =[u k (0),u k (1),…,u k(N-1)] T , then the linear discrete system in the form of equation (4) is translated into the following vector form:

[0023] y k =Hu k +Fx k (0)+R (5)

[0024] Among them, x k (0) is an arbitrarily chosen but bounded initial state of the system, H∈R N×N , F∈R N×2 and R∈R N×1 They are the parameter matrices formed by A, B, and C respectively:

[0025]

[0026] Among them, A 2 ,A 3 ,…A N Respectively represent the multiplication of 2, 3, ... N matrices A;

[0027] 2.5 To facilitate the estimation of unknown parameters, define the matrix P = [R, F, H] ∈ R N×(3+N) represents the matrix with unknown information, is a vector of known information; let (F v ) T and (R v ) T Represents the v-th row vector of matrices F and R, and replaces H and u k Using vectors H of different dimensions v ∈R v and Indicates; will (H v ) Τ Expressed as (H v ) Τ =[h v1 ,h v2 ,…,h vv ]∈R 1×v , where h vv represents the element corresponding to the vth row and vth column of H; similarly, the matrix P and the vector ψ k Described as vectors P of different dimensions v =[R v ,F v ,H v ]and Form, formula (5) is converted into the following vector form

[0028]

[0029] 2.6 For the actual application scenarios of permanent magnet synchronous motors, each system parameter must be bounded. Therefore, the parameters H, F, R, H defined in equations (5) and (6) are v , F v and R v It is bounded;

[0030] Step 3, design an adaptive iterative learning controller;

[0031] Step 4: Design an event trigger mechanism to realize data-driven adaptive iterative learning control of permanent magnet synchronous motor.

[0032] Further, the process of step 3 is as follows:

[0033] 3.1 For the parameter vector (P v ) T (v∈{1,2,…,N}) is unknown, use Represents a vector (P v ) T The estimated value at the kth iteration; get the output y of the system at the kth iteration k (v) The corresponding estimated value The form is as follows:

[0034]

[0035] 3.2 To minimize the estimation error The value of , construct the following optimization objective function:

[0036]

[0037] The objective function (8) is in the variable The gradient at is:

[0038]

[0039] Therefore, based on the gradient technique, the parameter vector The adaptive update algorithm is constructed as follows:

[0040]

[0041] Among them, the search step length μ>0 is a weighting factor;

[0042] 3.3 To make the parameter estimation algorithm have a stronger ability to track time-varying parameters and ensure that the parameter matrix The diagonal elements of Construct a reset algorithm. For any small η>0, if Then

[0043]

[0044] 3.4 Definition of signal y d (v) is the bounded desired speed of the permanent magnet synchronous motor when it moves, then e k+1 (v) = y d (v)-y k+1 (v) represents the error between the desired permanent magnet synchronous motor speed and the actual speed generated when the permanent magnet synchronous motor moves; according to (H v ) Τ The definition of Indicates (H v ) Τ The vector is composed of the first v-1 elements of . According to the error information and parameter estimation information, the adaptive iterative learning controller of the permanent magnet synchronous motor is designed as:

[0045]

[0046] Among them, γ k (v) is the learning gain that changes over time and iterations, designed as w>0 is an adjustable parameter;

[0047] Furthermore, the process of step 4 is as follows:

[0048] 4.1 Define the iterative sequence of event triggering as k s ,s=1,2,…, the input when triggering is The trigger error is Then for input u k (v) The designed triggering conditions are as follows:

[0049]

[0050] in, Λ1>0 and Λ2>1 are both bounded normal numbers.

[0051] 4.2 The data-driven adaptive iterative learning control method of permanent magnet synchronous motor based on event trigger mechanism is described as follows:

[0052]

[0053] in,

[0054] Most of the existing event trigger strategies are based on model-based control methods, while the strategy proposed in this paper is a data-driven control scheme, which means that the design and analysis process of the controller is independent of the permanent magnet synchronous motor system model information. In addition, according to the trigger condition, when the speed tracking error e of the permanent magnet synchronous motor is k (v) Especially when the system is small, the use of event trigger mechanism can avoid the permanent magnet synchronous motor from generating "unnecessary" frequent data transmission and prevent the system from triggering infinitely.

[0055] The present invention designs a data-driven adaptive iterative learning control method for a permanent magnet synchronous motor based on an event trigger mechanism. While reducing the influence of precise model information on the performance of the permanent magnet synchronous motor system, the proposed event trigger mechanism can effectively reduce the requirements of the permanent magnet synchronous motor controller on the system hardware performance and achieve good control of the permanent magnet synchronous motor system.

[0056] Technical concept of the present invention: For a permanent magnet synchronous motor system with unknown model parameter information, the present invention adopts the optimization objective function technology to process the unknown parameter items in the permanent magnet synchronous motor system, and then designs a parameter adaptive update algorithm, and designs an adaptive iterative learning controller based on the estimated parameter information and the input and output data of the system. The controller can ensure that the permanent magnet synchronous motor has a fast convergence capability, while effectively reducing the controller's dependence on accurate model information. In addition, an event trigger mechanism is designed to reduce the requirements of the permanent magnet synchronous motor controller on the system hardware performance. While the permanent magnet synchronous motor system can achieve a better control effect, it can also reduce the number of controller updates.

[0057] The beneficial effects of the present invention are as follows: 1. For systems with repetitive motion characteristics such as permanent magnet synchronous motors, the permanent magnet synchronous motors are taken as the controlled objects to achieve rapid tracking of the desired speed of the permanent magnet synchronous motors, reduce the dependence of the permanent magnet synchronous motor control method on accurate model information, and effectively save control resources; 2. Under the premise of ensuring that the tracking error of the permanent magnet synchronous motor system converges to near zero within a limited number of iterations when there is uncertainty in the parameters of the permanent magnet synchronous motor system, the number of controller updates can be reduced and the requirements for the system hardware performance can be reduced; at the same time, the controller is designed without relying on the specific parameters of the dynamic model, and only the input and output measurement data generated by the dynamic model are used to reduce the difficulty of model decoupling, which is more conducive to practical engineering promotion. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 is a control flow chart of the present invention;

[0059] Figure 2 Output tracking of different iteration times when the event triggering mechanism of the present invention is added;

[0060] Figure 3 The change of tracking error at different iteration times when the event trigger mechanism of the present invention is added;

[0061] Figure 4 is the number of event triggers at different sampling times of the present invention;

[0062] Figure 5 Detailed event triggering scenarios at different sampling times of the present invention. DETAILED DESCRIPTION

[0063] The present invention will be further described below in conjunction with the accompanying drawings.

[0064] Reference Figure 1 to Figure 5 , a data-driven adaptive iterative learning control method for permanent magnet synchronous motor based on event trigger mechanism, Figure 1 The model block diagram of the permanent magnet synchronous motor disclosed in the present invention is shown. The controller output of the kth batch is u k (t), and applying it to the permanent magnet synchronous motor can obtain the actual output y of the kth batch of the system k (t), and compare it with the set expected value y stored in the expected trajectory memory d (t) can be compared to obtain the actual tracking error e k (t). If the system does not reach the set maximum number of iterations, the error information and parameter estimation information are passed to the adaptive iterative learning controller to generate new outputs and determine whether the event triggering conditions are met. If the conditions are met, the newly generated controller output is used as the next batch of controller outputs u k+1 (t), if the condition is not met, the original saved controller output u k (t) Enter the next batch. This cycle runs until the system reaches the set maximum number of iterations, then stops iterating.

[0065] A data-driven adaptive iterative learning control method for a permanent magnet synchronous motor based on an event trigger mechanism comprises the following steps:

[0066] Step 1: Establish the dynamic model of permanent magnet synchronous motor:

[0067]

[0068] Wherein, t represents the time point, p(t) represents the rotor angular position of the permanent magnet synchronous motor, and w(t) represents the angular velocity of the permanent magnet synchronous motor during movement; and represent the first-order derivatives of p(t) and w(t), respectively; B is the viscous friction coefficient, J is the moment of inertia, i q is the q-axis stator current, n p is the number of pole pairs of the permanent magnet synchronous motor, ψf Represents magnetic flux, T L represents the load torque; in order to facilitate the design of the subsequent controller, let u(t) = i q , represents the input value of the system;

[0069] Step 2: Determine the discrete state space model of the permanent magnet synchronous motor system. The process is as follows:

[0070] 2.1 By selecting the sampling period T s , and use the Euler method to calculate the continuous system model (1) and Discretize it and get:

[0071]

[0072] 2.2 Substituting equation (2) into equation (1), after mathematical transformation, the discrete system model expression of the permanent magnet synchronous motor in motion is obtained as follows:

[0073]

[0074] Where v∈T N is a discrete time point; T N ={0,1,…,N}, N is a finite positive integer;

[0075] 2.3 Define the position p(v) and speed w(v) of the permanent magnet synchronous motor as the state variable x(v), expressed as: x(v) = [p(v) w(v)] Τ , the input variable of the permanent magnet synchronous motor system is defined as the current u(v), and the output variable is the angular velocity w(v); considering that the above permanent magnet synchronous motor system is repeatedly operated in a finite time, the permanent magnet synchronous motor model described by equation (3) is expressed as the following discrete state space model:

[0076]

[0077] Where k∈N + Indicates the number of system iterations, u k (v)∈R and y k (v)∈R represent the bounded single input and single output vectors of the system respectively; x k (v)∈R 2 Represents the state vector of the system; A, B, C and D are parameter matrices with appropriate dimensions, which contain various model information of the permanent magnet synchronous motor and satisfy CB≠0, in the following form:

[0078]

[0079] In order to reduce the dependence on physical model information in the practical application of permanent magnet synchronous motor, the present invention considers the situation that the information of parameter matrices A, B, C and D is unknown or incompletely known, and only uses the input and output data information in formula (4) to design the controller to realize data-driven control;

[0080] 2.4 To facilitate the design of subsequent controllers, input u k (v), output y k (v) Expressed in supervector form: y k =[y k (1),y k (2),…,y k (N)] T ,u k =[u k (0),u k (1),…,u k (N-1)] T , then the linear discrete system in the form of equation (4) is translated into the following vector form:

[0081] y k =Hu k +Fx k (0)+R (5)

[0082] Among them, x k (0) is an arbitrarily chosen but bounded initial state of the system, H∈R N×N , F∈R N×2 and R∈R N×1 They are the parameter matrices formed by A, B, and C respectively:

[0083]

[0084] Among them, A 2 ,A 3 ,…A N Respectively represent the multiplication of 2, 3, ... N matrices A;

[0085] 2.5 To facilitate the estimation of unknown parameters, define the matrix P = [R, F, H] ∈ R N×(3+N) represents the matrix with unknown information, is a vector of known information; let (F v ) T and (R v ) T Represents the v-th row vector of matrices F and R, and replaces H and u k Using vectors H of different dimensions v ∈R v and In particular, (H v ) ΤExpressed as (H v ) Τ =[h v1 ,h v2 ,…,h vv ]∈R 1×v , where h vv represents the element corresponding to the vth row and vth column of H; similarly, the matrix P and the vector ψ k Described as vectors P of different dimensions v =[R v ,F v ,H v ]and Form, formula (5) is converted into the following vector form

[0086]

[0087] 2.6 For the actual application scenarios of permanent magnet synchronous motors, each system parameter must be bounded. Therefore, the parameters H, F, R, H defined in equations (5) and (6) are v , F v and R v It is bounded;

[0088] Step 3: Design an adaptive iterative learning controller. The process is as follows:

[0089] 3.1 For the parameter vector (P v ) T (v∈{1,2,…,N}) is unknown, use Represents a vector (P v ) T The estimated value at the kth iteration; further, the output y of the system at the kth iteration is obtained k (v) The corresponding estimated value The form is as follows:

[0090]

[0091] 3.2 To minimize the estimation error The value of , construct the following optimization objective function:

[0092]

[0093] The objective function (8) is in the variable The gradient at is:

[0094]

[0095] Therefore, based on the gradient technique, the parameter vector The adaptive update algorithm is constructed as follows:

[0096]

[0097] Among them, the search step length μ>0 is a weighting factor;

[0098] 3.3 To make the above parameter estimation algorithm have a stronger ability to track time-varying parameters and ensure that the parameter matrix The diagonal elements of Construct a reset algorithm: For any small η>0, if Then

[0099]

[0100] 3.4 Definition of signal y d (v) is the bounded desired speed of the permanent magnet synchronous motor when it moves, then e k+1 (v) = y d (v)-y k+1 (v) represents the error between the desired permanent magnet synchronous motor speed and the actual speed generated when the permanent magnet synchronous motor moves; according to (H v ) Τ The definition of Indicates (H v ) Τ The vector is composed of the first v-1 elements of . According to the error information and parameter estimation information, the adaptive iterative learning controller of the permanent magnet synchronous motor is designed as:

[0101]

[0102] Among them, γ k (v) is the learning gain that changes over time and iterations, designed as w>0 is an adjustable parameter;

[0103] Step 4, design of event triggering mechanism, the process is as follows:

[0104] 4.1 Define the iterative sequence of event triggering as k s ,s=1,2,…, the input when triggering is The trigger error is Then for input u k (v) The designed triggering conditions are as follows:

[0105]

[0106] in, Λ1>0 and Λ2>1 are both bounded normal numbers.

[0107] 4.2 The data-driven adaptive iterative learning control method of permanent magnet synchronous motor based on event trigger mechanism is described as follows:

[0108]

[0109] in,

[0110] In order to verify the effectiveness of the proposed method, the present invention conducts a simulation experiment on the control effect of the data-driven adaptive iterative learning control method of the permanent magnet synchronous motor based on the event trigger mechanism shown in equations (14) and (15); the control parameters in the dynamic model of the permanent magnet synchronous motor shown in equation (1) are set as: B = 0.12, J = 0.0009 kg·m 2 , n p =4,ψ f =0.00162Wb, T L =0.0009N·m,

[0111]

[0112] The simulation time is set to t = 1s, and the sampling time is set to T s =0.01s, then N = 100, and the discrete state expression of the system can be described as:

[0113]

[0114] Set the initial state in the permanent magnet synchronous motor experiment to x k (0) = x0 = 0.01rand(2,1), initial input u1(v) = 0, the expected output trajectory is:

[0115] y d (v) = sin0.05v (17)

[0116] Select the parameters Λ1=0.01 and Λ2=2 in the event trigger mechanism (13); select the parameter w=0.001 in the controller (12), the parameter μ=0.01 in the parameter adaptive update algorithm (14), and the parameter estimation initial value Select as:

[0117]

[0118] Figure 2 to Figure 5 This is a comparison chart of the effects of the data-driven adaptive iterative learning control method for permanent magnet synchronous motors based on an event trigger mechanism of the present invention. Figure 2 and Figure 3It can be seen that the actual speed of the permanent magnet synchronous motor gradually tracks the expected speed as the number of iterations increases; the maximum tracking error decreases and gradually tends to zero at the 50th iteration. This shows that the data-driven adaptive iterative learning control method of the present invention can achieve good performance of the permanent magnet synchronous motor by only using the input and output data generated by the permanent magnet synchronous motor system without relying on the precise model information of the permanent magnet synchronous motor. Figure 4 and Figure 5 are the sampling frequencies under the event trigger mechanism. Figure 4 It can be clearly seen that the number of iterative updates corresponding to the trigger at 100 sampling points is significantly reduced, which indicates that the input update frequency is significantly reduced. Figure 5 Four specific sampling time points are selected to show the specific event triggering situation. The vertical axis 1 indicates that the event is triggered, the input is iteratively updated, and a new value is generated; 0 indicates that the event is not triggered, and the input maintains the same value as the previous iteration. Obviously, it can be seen that the update frequency of the input is greatly reduced. From the above simulation results, it can be seen that when the controller is updated according to the event trigger mechanism, the data-driven adaptive iterative learning control method proposed in the present invention enables the permanent magnet synchronous motor system to achieve rapid tracking of the desired speed, while reducing the permanent magnet synchronous motor control method's dependence on precise model information, and can reduce the number of controller updates, ensuring effective saving of control resources.

[0119] The above is a simulation result given by the present invention. Obviously, the present invention is not limited to the above examples, and various modifications can be made to implement it without departing from the basic spirit of the present invention and without exceeding the scope of the essential content of the present invention.

Claims

1. A data-driven adaptive iterative learning control method for a permanent magnet synchronous motor based on an event-triggered mechanism, characterized in that: The method comprises the following steps: Step 1: Establish the dynamic model of permanent magnet synchronous motor: Wherein, t represents the time point, p(t) represents the rotor angular position of the permanent magnet synchronous motor, and w(t) represents the angular velocity of the permanent magnet synchronous motor during movement; and represent the first-order derivatives of p(t) and w(t), respectively; B is the viscous friction coefficient, J is the moment of inertia, i q is the q-axis stator current, n p is the number of pole pairs of the permanent magnet synchronous motor, ψ f Represents magnetic flux, T L represents the load torque; in order to facilitate the design of the subsequent controller, let u(t) = i q , represents the input value of the system; Step 2: Determine the discrete state space model of the permanent magnet synchronous motor system. The process is as follows: 2.1 By selecting the sampling period T s , and use the Euler method to calculate the continuous system model (1) and Discretize it and get: 2.2 Substituting equation (2) into equation (1), after mathematical transformation, the discrete system model expression of the permanent magnet synchronous motor in motion is obtained as follows: in, v∈T N is a discrete time point; T N ={0,1,…,N}, N is a finite positive integer; 2.3 Define the position p(v) and speed w(v) of the permanent magnet synchronous motor as the state variable x(v), expressed as: x(v) = [p(v)w(v)] Τ , the input variable of the permanent magnet synchronous motor system is defined as the current u(v), and the output variable is the angular velocity w(v); considering that the permanent magnet synchronous motor system is repeatedly operated in a finite time, the permanent magnet synchronous motor model described by equation (3) is expressed as the following discrete state space model: Where k∈N + Indicates the number of system iterations, u k (v)∈R and y k (v)∈R represent the bounded single input and single output vectors of the system respectively; x k (v)∈R 2 Represents the state vector of the system; A, B, C and D are parameter matrices with appropriate dimensions, which contain various model information of the permanent magnet synchronous motor and satisfy CB≠0, in the following form: C＝[0 1], In the case where the information of parameter matrices A, B, C and D is unknown or incompletely known, only the input and output data information in equation (4) is used to design the controller to achieve data-driven control. 2.4 Input u k (v), output y k (v) Expressed in supervector form: y k =[y k (1),y k (2),…,y k (N)] T ,u k =[u k (0),u k (1),…,u k (N-1)] T , then the linear discrete system in the form of equation (4) is translated into the following vector form: the k =Hu k +Fx k (0)+R (5) Among them, x k (0) is an arbitrarily chosen but bounded initial state of the system, H∈R N×N , F∈R N×2 and R∈R N×1 They are the parameter matrices formed by A, B, and C respectively: Among them, A 2 ,A 3 ,…A N Respectively represent the multiplication of 2, 3, ... N matrices A; 2.5 Define the matrix P = [R, F, H] ∈ R N×(3+N) represents the matrix with unknown information, is a vector of known information; let (F v ) T and (R v ) T Represents the v-th row vector of matrices F and R, and replaces H and u k Using vectors H of different dimensions v ∈R v and Indicates; will (H v ) Τ Expressed as (H v ) Τ =[h v1 ,h v2 ,…,h vv ]∈R 1×v , where h vv represents the element corresponding to the vth row and vth column of H; similarly, the matrix P and the vector ψ k Described as vectors P of different dimensions v =[R v ,F v ,H v ]and Formula (5) is converted into the following vector form 2.6 For the actual application scenarios of permanent magnet synchronous motors, each system parameter must be bounded. Therefore, the parameters H, F, R, H defined in equations (5) and (6) are v , F v and R v It is bounded; Step 3, design an adaptive iterative learning controller; Step 4: Design an event trigger mechanism to realize data-driven adaptive iterative learning control of permanent magnet synchronous motor.

2. The data-driven adaptive iterative learning control method for a permanent magnet synchronous motor based on an event trigger mechanism according to claim 1, wherein the process of step 3 is as follows: 3.1 For the parameter vector (P v ) T (v∈{1,2,…,N}) is unknown, use Represents a vector (P v ) T The estimated value at the kth iteration; get the output y of the system at the kth iteration k (v) The corresponding estimated value The form is as follows: 3.2 To minimize the estimation error The value of , construct the following optimization objective function: The objective function (8) is in the variable The gradient at is: Therefore, based on the gradient technique, the parameter vector The adaptive update algorithm is constructed as follows: Among them, the search step length μ>0 is a weighting factor; 3.3 To make the parameter estimation algorithm have a stronger ability to track time-varying parameters and ensure that the parameter matrix The diagonal elements of Construct a reset algorithm: For any small η>0, if Then 3.4 Definition of signal y d (v) is the bounded desired speed of the permanent magnet synchronous motor when it moves, then e k+1 (v) = y d (v)-y k+1 (v) represents the error between the desired permanent magnet synchronous motor speed and the actual speed generated when the permanent magnet synchronous motor moves; according to (H v ) Τ The definition of Indicates (H v ) Τ The vector is composed of the first v-1 elements of . According to the error information and parameter estimation information, the adaptive iterative learning controller of the permanent magnet synchronous motor is designed as: Among them, γ k (v) is the learning gain that changes over time and iterations, designed as w>0 is an adjustable parameter; 3. The data-driven adaptive iterative learning control method for a permanent magnet synchronous motor based on an event-triggered mechanism according to claim 1 or 2, characterized in that: The process of step 4 is as follows: 4.1 Define the iterative sequence of event triggering as k s ,s=1,2,…, the input when triggering is The trigger error is Then for input u k (v) The designed triggering conditions are as follows: in, Λ1>0 and Λ2>1 are both bounded normal numbers. 4.2 The data-driven adaptive iterative learning control method of permanent magnet synchronous motor based on event trigger mechanism is described as follows: in,

Images