Parameter identification method and apparatus for permanent-magnet synchronous motor, and storage medium and device

Abstract

The present invention belongs to the technical field of parameter identification for permanent-magnet synchronous motors. Disclosed are a parameter identification method and apparatus for a permanent-magnet synchronous motor, and a storage medium and a device. The method comprises: using a variable step-size feedback neural network algorithm to solve a voltage equation and a d-q axis current discrete state equation, which take cross-coupled inductance into consideration, of a permanent-magnet synchronous motor in a dq coordinate system, so as to obtain intermediate parameters generated in the process of constructing the d-q axis current discrete state equation; and performing calculation on the basis of the intermediate parameters, so as to obtain the stator resistance, the d-axis inductance, the q-axis inductance, the cross-coupled inductance and the permanent-magnet flux linkage of the permanent-magnet synchronous motor, and thereby completing parameter identification. The present invention takes cross-coupled inductance into consideration and uses the variable step-size feedback neural network algorithm, such that the computational load is extremely small, the degree of complexity is reduced, rapid and effective identification can be realized, and errors can be reduced; in different convergence time periods, the algorithm adjusts the step size on the basis of the magnitude of steady-state errors, such that identification results approach stable values more quickly, thereby improving the speed and stability of convergence.

Description

A method, apparatus, storage medium, and device for parameter identification of a permanent magnet synchronous motor. Technical Field

[0001] This invention relates to a method, apparatus, storage medium, and device for identifying parameters of a permanent magnet synchronous motor, belonging to the field of permanent magnet synchronous motor parameter identification technology. Background Technology

[0002] Permanent magnet synchronous motors (PMSMs) have advantages such as small size, simple structure, and high power density, and have been widely used in industrial fields such as home appliances and electric vehicles in recent years. The efficient implementation of various control strategies for PMSMs is closely related to their own parameters. During actual operation, the motor's parameters are affected by the operating environment, load conditions, and harmonic interference, resulting in cross-coupled inductance and thus affecting the performance of the motor controller. Therefore, accurately identifying motor parameters is crucial for improving the stability of the motor control system during PMSM driver software development.

[0003] In existing technologies, algorithms such as recursive least squares, model reference adaptation, and extended Kalman filtering are commonly used for parameter identification.

[0004] The least squares method has advantages such as simplicity and ease of implementation, but it requires handling massive amounts of data. As the data grows, the recursive least squares method will experience a "data saturation" phenomenon, making newly collected data less effective in updating the parameter estimates. When the parameters change, the recursive least squares method will be unable to keep up with these changes, leading to failure in online parameter identification.

[0005] The recursive least squares method with a forgetting factor can reduce the impact of data saturation to some extent. However, the forgetting factor in this algorithm is a fixed value. When the value is too small, it will affect the robustness of the algorithm. When the value of the forgetting factor is too large, it will significantly affect the convergence speed and stability of the algorithm.

[0006] In model-referenced adaptive methods, the selection of the identifier parameters has a significant impact on the identification performance. Ignoring the underrank problem of the state equations can lead to the identification results getting trapped in local optima. Correct convergence of the parameters depends on the selection of their initial values. Using rated resistance and flux linkage for parameter identification introduces errors, resulting in low overall system recognition accuracy. Furthermore, adjustable adaptive models are complex and difficult to establish.

[0007] The system noise covariance matrix and measurement noise covariance matrix of EKF are difficult to select, further affecting the accuracy and speed of parameter identification. In addition, the algorithm iteration process requires a large number of matrix and vector operations, and precise preprocessing of the motor mathematical model is also required, resulting in a huge amount of computation during the identification process, and the calculation is very complex when there are many identification parameters.

[0008] Furthermore, existing permanent magnet synchronous motor parameter identification technologies do not consider the motor cross-coupling effect and do not perform parameter identification for cross-coupling inductance. Summary of the Invention

[0009] The purpose of this invention is to provide a method, device, storage medium and equipment for identifying parameters of permanent magnet synchronous motors, and to solve the problems of poor stability, slow convergence speed, large error and failure to consider the cross-coupling effect of motors in the prior art.

[0010] To achieve the above objectives, the present invention employs the following technical solution:

[0011] In a first aspect, the present invention provides a method for identifying parameters of a permanent magnet synchronous motor, comprising:

[0012] The voltage equation and dq-axis current discrete state equation of permanent magnet synchronous motor in dq coordinate system considering cross-coupled inductance are solved by using a variable step size feedback neural network algorithm, and intermediate parameters generated in the process of constructing the dq-axis current discrete state equation are obtained.

[0013] Based on the intermediate parameters, the stator resistance, d-axis inductance, q-axis inductance, cross-coupling inductance, and permanent magnet flux linkage of the permanent magnet synchronous motor are calculated, thus completing parameter identification.

[0014] Furthermore, the voltage equation is as follows:

[0015] Among them, u d It is the d-axis voltage, u q It is the q-axis voltage, i d It is the d-axis current, i q It is the q-axis current, R s It is the stator resistance, ω e It is the electric angular velocity, L d It is the d-axis inductance, L q It is the q-axis inductance, ψ f It is a permanent magnet flux linkage, L dq It is the first cross-coupled inductor, L qd It is the second cross-coupled inductor, and L dq =L qd .

[0016] Furthermore, the discrete state equation for the dq-axis current is obtained through the following method:

[0017] In the voltage equation, L is used dq Replace L qd Then, by rearranging terms and deriving, the current equation of the permanent magnet synchronous motor considering the cross-coupled inductance in the dq coordinate system is obtained;

[0018] The current equation is discretized using the Pad approximation method to obtain the discrete state equation of the dq-axis current.

[0019] Furthermore, the current equation is as follows:

[0020] Furthermore, the discrete state equation for the dq-axis current is:

[0021] Among them, β1, β2, β3, β4, and β5 are intermediate parameters generated during the discretization of the current equation using the Pad approximation method, i d(k+1) It is the actual value of the d-axis current of the permanent magnet synchronous motor at time k+1, i d(k) i is the actual value of the d-axis current of the permanent magnet synchronous motor at time k. d(k+2) It is the actual value of the d-axis current of the permanent magnet synchronous motor at time k+2, i q(k+1) It is the actual value of the q-axis current of the permanent magnet synchronous motor at time k+1, i q(k) It is the actual value of the q-axis current of the permanent magnet synchronous motor at time k, i q(k+2) This is the actual value of the q-axis current of the permanent magnet synchronous motor at time k+2, u d(k+1) This is the actual value of the d-axis voltage of the permanent magnet synchronous motor at time k+1, u d(k) U is the actual value of the d-axis voltage of the permanent magnet synchronous motor at time k. q(k+1) It is the actual value of the q-axis voltage of the permanent magnet synchronous motor at time k+1, u q(k) ω is the actual value of the q-axis voltage of the permanent magnet synchronous motor at time k. e(k+1) It is the actual value of the electric angular velocity of the permanent magnet synchronous motor at time k+1, ω e(k) T is the actual value of the electric angular velocity of the permanent magnet synchronous motor at time k. s is the sampling period, and k is the sampling time sequence number.

[0022] Furthermore, the stator resistance, d-axis inductance, q-axis inductance, cross-coupling inductance, and permanent magnet flux linkage of the permanent magnet synchronous motor are calculated based on the intermediate parameters using the following formula:

[0023] Furthermore, the activation function of the neural network in the variable step-size feedback neural network algorithm is: O(k) = ∑W i X i ;

[0024] Where O(k) is kT s The output of the time-matter neural network, W i X is the weight of the i-th layer of the neural network. i It is the input to the i-th layer of the neural network;

[0025] In solving the voltage equation and dq-axis current discrete state equation of a permanent magnet synchronous motor in the dq coordinate system considering cross-coupling effects using a variable step-size feedback neural network algorithm, the weights are adjusted using a weight adjustment algorithm, the expression of which is:

[0026] Where W(k+1) is (k+1)T s The weights of the neural network at time step kT, W(k) is... s The weights of the neural network at time step η are the step size for weight adjustment, and X(k) is the weight of the kT neural network. s The input to the neural network at time step k, d(k) is kT s The target output of the time-limited neural network, a and b are pre-set adjustable coefficients, ε(k) represents the difference between d(k) and O(k), and e is the natural constant;

[0027] Wherein, η needs to satisfy 0 < 2η|X(k)| 2 <1.

[0028] Secondly, the present invention provides a parameter identification device for a permanent magnet synchronous motor, comprising:

[0029] The equation solving module is configured to use a variable step size feedback neural network algorithm to solve the voltage equation and the discrete state equation of the dq axis current of the permanent magnet synchronous motor in the dq coordinate system, considering the cross-coupled inductance, and obtain the intermediate parameters generated in the process of constructing the discrete state equation of the dq axis current.

[0030] The parameter identification module is configured to calculate the stator resistance, d-axis inductance, q-axis inductance, cross-coupled inductance, and permanent magnet flux linkage of the permanent magnet synchronous motor based on the intermediate parameters, thereby completing parameter identification.

[0031] Thirdly, the present invention provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implements the steps of the permanent magnet synchronous motor parameter identification method described in any one of the first aspects.

[0032] Fourthly, the present invention provides a computer device, comprising:

[0033] Memory, used to store computer programs / instructions;

[0034] A processor is configured to execute the computer program / instructions to implement the steps of the permanent magnet synchronous motor parameter identification method as described in any one of the first aspects.

[0035] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0036] This invention provides a method, apparatus, storage medium, and device for identifying parameters of a permanent magnet synchronous motor (PMSM). Considering the cross-coupled inductance during PMSM operation, a variable step-size feedback neural network algorithm is used to identify PMSM parameters. The computational load is very small, and the complexity is greatly reduced compared to traditional algorithms. It can quickly and effectively identify multiple parameters of the PMSM to ensure the control performance of the motor and reduce errors. The variable step-size feedback neural network algorithm adjusts the identification step size according to the steady-state error at different convergence time periods, thereby making the identification results approach the stable value more quickly and improving the convergence speed and stability. Attached Figure Description

[0037] Figure 1 is a flowchart of a method for identifying parameters of a permanent magnet synchronous motor according to an embodiment of the present invention;

[0038] Figure 2 is a vector control structure diagram of a permanent magnet synchronous motor provided in an embodiment of the present invention;

[0039] Figure 3 is a schematic diagram of the identification principle based on the variable step size feedback neural network algorithm provided in an embodiment of the present invention;

[0040] Figure 4 is a schematic diagram of the neural network structure provided in an embodiment of the present invention. Detailed Implementation

[0041] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to illustrate the technical solution of the present invention more clearly, and should not be used to limit the scope of protection of the present invention.

[0042] Example 1

[0043] As shown in Figure 1, the present invention provides a method for identifying parameters of a permanent magnet synchronous motor, comprising the following steps:

[0044] S1. The voltage equation and dq-axis current discrete state equation of the permanent magnet synchronous motor in the dq coordinate system considering the cross-coupled inductance are solved by using the variable step size feedback neural network algorithm, and the intermediate parameters generated in the process of constructing the dq-axis current discrete state equation are obtained.

[0045] According to the control structure diagram in Figure 2:

[0046] Typically, the voltage equations and flux linkage equations for an internal permanent magnet synchronous motor in a two-phase synchronous rotating coordinate system (dq coordinate system) are as follows:

[0047] However, during actual motor operation, the common magnetic circuit of the quadrature and direct axes causes magnetic circuit crossing, so the saturation levels of the quadrature and direct axis magnetic fields affect each other, further generating cross-coupled inductance L. dq and L qdWhen considering the effect of the motor's cross-coupling inductance, its motor voltage equation can be expressed as:

[0048] Because of L dq Approximately equal to L qd Use L uniformly dq This indicates that the current equation for the PMSM can be further derived as follows:

[0049] Among them, u d It is the d-axis voltage, u q It is the q-axis voltage, i d It is the d-axis current, i q It is the q-axis current, R s It is the stator resistance, ω e It is the electric angular velocity, L d It is the d-axis inductance, L q It is the q-axis inductance, ψ f It is a permanent magnet flux linkage, L dq It is the first cross-coupled inductor, L qd It is the second cross-coupled inductor, and L dq =L qd .

[0050] Discretizing the current equations of the PMSM using the Pad approximation method yields the discrete state equations for the dq-axis currents of the PMSM:

[0051] Among them, β1, β2, β3, β4, and β5 are intermediate parameters generated during the discretization of the current equation using the Pad approximation method, i d(k+1) It is the actual value of the d-axis current of the permanent magnet synchronous motor at time k+1, i d(k) i is the actual value of the d-axis current of the permanent magnet synchronous motor at time k. d(k+2) It is the actual value of the d-axis current of the permanent magnet synchronous motor at time k+2, i q(k+1) It is the actual value of the q-axis current of the permanent magnet synchronous motor at time k+1, i q(k) It is the actual value of the q-axis current of the permanent magnet synchronous motor at time k, i q(k+2) This is the actual value of the q-axis current of the permanent magnet synchronous motor at time k+2, u d(k+1) This is the actual value of the d-axis voltage of the permanent magnet synchronous motor at time k+1, u d(k) U is the actual value of the d-axis voltage of the permanent magnet synchronous motor at time k. q(k+1) It is the actual value of the q-axis voltage of the permanent magnet synchronous motor at time k+1, u q(k) ω is the actual value of the q-axis voltage of the permanent magnet synchronous motor at time k. e(k+1) It is the actual value of the electric angular velocity of the permanent magnet synchronous motor at time k+1, ω e(k)T is the actual value of the electric angular velocity of the permanent magnet synchronous motor at time k. s is the sampling period, and k is the sampling time sequence number.

[0052] in:

[0053] This embodiment employs a variable step-size feedback neural network algorithm. The step size is only functionally related to the current steady-state error, thus avoiding the influence of the accumulation of iteration error on the step size. This algorithm has a fast convergence speed and can achieve a small steady-state error.

[0054] As shown in Figure 4, the variable step size feedback neural network algorithm includes an activation function, which can be expressed as: O(k)=∑W i X i ;

[0055] Where O(k) is kT s The output of the time-matter neural network, W i X is the weight of the i-th layer of the neural network. i It is the input to the i-th layer of the neural network.

[0056] A variable step-size feedback neural network algorithm is adopted, and a corresponding functional relationship is established between the step size and the square of the instantaneous error, so that the step size gradually decreases as the steady-state error decreases.

[0057] In solving the voltage equation and dq-axis current discrete state equation of a permanent magnet synchronous motor in the dq coordinate system considering cross-coupling effects using a variable step-size feedback neural network algorithm, the weights are adjusted using a weight adjustment algorithm, as shown in Figure 3 (LMS, Least Mean Squares). The expression for the weight adjustment algorithm is:

[0058] Where W(k+1) is (k+1)T s The weights of the neural network at time step kT, W(k) is... s The weights of the neural network at time step η are the step size for weight adjustment, and X(k) is the weight of the kT neural network. s The input to the neural network at time step k, d(k) is kT s The target output of the time-limited neural network, a and b are pre-set adjustable coefficients, ε(k) represents the difference between d(k) and O(k), and e is the natural constant;

[0059] To ensure the convergence of the algorithm, η needs to satisfy 0 < 2η|X(k)| 2 <1.

[0060] Define the input matrix for the d-axis. The output matrix y(k+1) is as follows:

[0061] Furthermore, we can obtain:

[0062] Among them, θ(k)=[β1β2β3β4] T This is the matrix of unknown parameters of the system.

[0063] Define the input matrix φ of the d-axis T The (k+1) and output matrix z(k+1) are respectively:

[0064] Furthermore, we can obtain: z(k+1)=φ T (k+1)δ(k);

[0065] Among them, δ(k)=[β1β2β3β4β5] T Given the unknown parameter matrix of the system, the values ​​of β1 to β5 can be obtained by solving the Pad approximation equation using a variable step size feedback neural network algorithm, thereby identifying the parameters of the permanent magnet synchronous motor.

[0066] S2. Based on the intermediate parameters, the stator resistance, d-axis inductance, q-axis inductance, cross-coupling inductance, and permanent magnet flux linkage of the permanent magnet synchronous motor are calculated to complete parameter identification.

[0067] Step S2 is performed using the following formula:

[0068] Example 2

[0069] This invention provides a parameter identification device for a permanent magnet synchronous motor, comprising:

[0070] The equation solving module is configured to use a variable step size feedback neural network algorithm to solve the voltage equation and the discrete state equation of the dq axis current of the permanent magnet synchronous motor in the dq coordinate system, considering the cross-coupled inductance, and obtain the intermediate parameters generated in the process of constructing the discrete state equation of the dq axis current.

[0071] The parameter identification module is configured to calculate the stator resistance, d-axis inductance, q-axis inductance, cross-coupled inductance, and permanent magnet flux linkage of the permanent magnet synchronous motor based on the intermediate parameters, thereby completing parameter identification.

[0072] Example 3

[0073] This invention provides a computer-readable storage medium storing a computer program / instructions thereon, which, when executed by a processor, implements the steps of the permanent magnet synchronous motor parameter identification method provided in Embodiment 1:

[0074] The voltage equation and dq-axis current discrete state equation of permanent magnet synchronous motor in dq coordinate system considering cross-coupled inductance are solved by using a variable step size feedback neural network algorithm, and intermediate parameters generated in the process of constructing the dq-axis current discrete state equation are obtained.

[0075] Based on the intermediate parameters, the stator resistance, d-axis inductance, q-axis inductance, cross-coupling inductance, and permanent magnet flux linkage of the permanent magnet synchronous motor are calculated, thus completing parameter identification.

[0076] Example 4

[0077] This invention provides a computer device, comprising:

[0078] Memory, used to store computer programs / instructions;

[0079] A processor is configured to execute the computer program / instructions to implement the steps of the permanent magnet synchronous motor parameter identification method provided in Embodiment 1:

[0080] The voltage equation and dq-axis current discrete state equation of permanent magnet synchronous motor in dq coordinate system considering cross-coupled inductance are solved by using a variable step size feedback neural network algorithm, and intermediate parameters generated in the process of constructing the dq-axis current discrete state equation are obtained.

[0081] Based on the intermediate parameters, the stator resistance, d-axis inductance, q-axis inductance, cross-coupling inductance, and permanent magnet flux linkage of the permanent magnet synchronous motor are calculated, thus completing parameter identification.

[0082] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0083] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in one or more flowchart illustrations and / or one or more block diagrams.

[0084] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means that implement the functions specified in one or more flowcharts and / or one or more block diagrams.

[0085] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, such that the instructions, which execute on the computer or other programmable apparatus, provide steps for implementing the functions specified in one or more flowcharts and / or one or more block diagrams.

[0086] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for identifying parameters of a permanent magnet synchronous motor, characterized in that, include: The voltage equation and dq-axis current discrete state equation of permanent magnet synchronous motor in dq coordinate system considering cross-coupled inductance are solved by using a variable step size feedback neural network algorithm, and intermediate parameters generated in the process of constructing the dq-axis current discrete state equation are obtained. Based on the intermediate parameters, the stator resistance, d-axis inductance, q-axis inductance, cross-coupling inductance, and permanent magnet flux linkage of the permanent magnet synchronous motor are calculated, thus completing parameter identification.

2. The method for identifying parameters of a permanent magnet synchronous motor according to claim 1, characterized in that, The voltage equation is as follows: Among them, u d It is the d-axis voltage, u q It is the q-axis voltage, i d It is the d-axis current, i q It is the q-axis current, R s It is the stator resistance, ω e It is the electric angular velocity, L d It is the d-axis inductance, L q It is the q-axis inductance, ψ f It is a permanent magnet flux linkage, L dq It is the first cross-coupled inductor, L qd It is the second cross-coupled inductor, and L dq =L qd .

3. The method for identifying parameters of a permanent magnet synchronous motor according to claim 2, characterized in that, The discrete state equations for the dq-axis currents are obtained through the following method: In the voltage equation, L is used dq Replace L qd Then, by rearranging terms and deriving, the current equation of the permanent magnet synchronous motor considering the cross-coupled inductance in the dq coordinate system is obtained; The current equation is discretized using the Pad approximation method to obtain the discrete state equation of the dq-axis current.

4. The method for identifying parameters of a permanent magnet synchronous motor according to claim 3, characterized in that, The current equation is as follows:

5. The method for identifying parameters of a permanent magnet synchronous motor according to claim 3, characterized in that, The discrete state equations of the d-q axis current are: i d(k+1) = β1i d(k) + β2(u d(k+1) + u d(k) ) + β3(ω e(k+1) i q(k+1) + ω e(k) i q(k) ) + β4(T s ω e(k+1) i d(k+1) + T s ω e(k) i d(k) + i q(k) - i q(k+2) ); i q(k+1) = β1i q(k) + β2[u q(k+1) + u q(k) - β3[ω e(k+1) i d(k+1) + ω e(k) i d(k) - β4[T s ω e(k+1) i q(k+1) + T s ω e(k) i q(k) + i d(k) - i d(k+2) + β5[ω e(k+1) + ω e(k) ; Among them, β1, β2, β3, β4, and β5 are intermediate parameters generated during the discretization of the current equation using the Pad approximation method, i d(k+1) It is the actual value of the d-axis current of the permanent magnet synchronous motor at time k+1, i d(k) i is the actual value of the d-axis current of the permanent magnet synchronous motor at time k. d(k+2) It is the actual value of the d-axis current of the permanent magnet synchronous motor at time k+2, i q(k+1) It is the actual value of the q-axis current of the permanent magnet synchronous motor at time k+1, i q(k) It is the actual value of the q-axis current of the permanent magnet synchronous motor at time k, i q(k+2) This is the actual value of the q-axis current of the permanent magnet synchronous motor at time k+2, u d(k+1) This is the actual value of the d-axis voltage of the permanent magnet synchronous motor at time k+1, u d(k) U is the actual value of the d-axis voltage of the permanent magnet synchronous motor at time k. q(k+1) It is the actual value of the q-axis voltage of the permanent magnet synchronous motor at time k+1, u q(k) ω is the actual value of the q-axis voltage of the permanent magnet synchronous motor at time k. e(k+1) It is the actual value of the electric angular velocity of the permanent magnet synchronous motor at time k+1, ω e(k) T is the actual value of the electric angular velocity of the permanent magnet synchronous motor at time k. s is the sampling period, and k is the sampling time sequence number.

6. The method for identifying parameters of a permanent magnet synchronous motor according to claim 5, characterized in that, The stator resistance, d-axis inductance, q-axis inductance, cross-coupling inductance, and permanent magnet flux linkage of the permanent magnet synchronous motor are calculated based on the intermediate parameters using the following formulas:

7. The method for identifying parameters of a permanent magnet synchronous motor according to claim 1, characterized in that, The activation function of the neural network in the variable step-size feedback neural network algorithm is: O(k) = ∑W i X i ; Where O(k) is kT s The output of the time-matter neural network, W i X is the weight of the i-th layer of the neural network. i It is the input to the i-th layer of the neural network; In solving the voltage equation and dq-axis current discrete state equation of a permanent magnet synchronous motor in the dq coordinate system considering cross-coupling effects using a variable step-size feedback neural network algorithm, the weights are adjusted using a weight adjustment algorithm, the expression of which is: Where W(k+1) is (k+1)T s The weights of the neural network at time step kT, W(k) is... s The weights of the neural network at time step η are the step size for weight adjustment, and X(k) is the weight of the kT neural network. s The input to the neural network at time step k, d(k) is kT s The target output of the time-limited neural network, a and b are pre-set adjustable coefficients, ε(k) represents the difference between d(k) and O(k), and e is the natural constant; Wherein, η needs to satisfy 0 < 2η|X(k)| 2 <1.

8. A parameter identification device for a permanent magnet synchronous motor, characterized in that, include: The equation solving module is configured to use a variable step size feedback neural network algorithm to solve the voltage equation and the discrete state equation of the dq axis current of the permanent magnet synchronous motor in the dq coordinate system, considering the cross-coupled inductance, and obtain the intermediate parameters generated in the process of constructing the discrete state equation of the dq axis current. The parameter identification module is configured to calculate the stator resistance, d-axis inductance, q-axis inductance, cross-coupled inductance, and permanent magnet flux linkage of the permanent magnet synchronous motor based on the intermediate parameters, thereby completing parameter identification.

9. A computer-readable storage medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instruction is executed by the processor, it implements the steps of the permanent magnet synchronous motor parameter identification method according to any one of claims 1-7.

10. A computer device, characterized in that, include: Memory, used to store computer programs / instructions; A processor for executing the computer program / instructions to implement the steps of the permanent magnet synchronous motor parameter identification method according to any one of claims 1-7.

Images