Dead-beat current prediction control method based on high-order terminal sliding mode observer

Abstract

The invention discloses a deadbeat current prediction control mode based on a high-order terminal sliding mode observer. On the basis of traditional dead-beat current prediction control, aiming at the defect that the traditional dead-beat current prediction control is influenced by motor model lumped disturbance when being applied to a permanent magnet synchronous motor current loop, a high-order terminal sliding mode observer is introduced into the traditional current prediction control for optimization. A motor model lumped disturbance value is estimated and compensated through a high-order terminal sliding-mode observer, buffeting caused by a traditional sliding-mode observer is reduced, rapid convergence of a current predicted value is achieved, and a steady-state static error of motor current is eliminated. The motor current loop applied to the method has excellent dynamic and steady-state performance for given current.

Description

Technical Field

[0001] The present invention relates to the field of permanent magnet synchronous motor control, and in particular to a deadbeat current prediction control method based on a high-order terminal sliding mode observer. Background Art

[0002] Permanent magnet synchronous motors (PMSMs) are widely used in servo systems due to their simple structure, high power density, and ease of control. As the inner loop of a PMSM servo control system, the control performance of the current loop directly impacts the system's dynamic response and steady-state accuracy. Therefore, the design and optimization of the current loop has become a research hotspot in the field of motor control.

[0003] Deadbeat current predictive control (DCC) calculates the required voltage reference value based on the motor's mathematical model. It offers advantages such as fast response and ease of digital implementation. However, in practical applications, the model not only contains unmodeled dynamics, but also model parameters that change with changing operating conditions. Consequently, there is a certain discrepancy between the voltage reference value calculated using the motor model and the actual voltage required by the motor, resulting in steady-state current tracking errors. In severe cases, the current can oscillate significantly, degrading system control performance. To address these issues, it is necessary to improve traditional deadbeat current predictive control to eliminate the effects of model mismatch. Summary of the Invention

[0004] The purpose of this invention is to eliminate the drawback of traditional deadbeat technology, described in the background art, that is significantly affected by the model, and to provide a deadbeat current predictive control method based on a high-order terminal sliding mode observer. This invention combines deadbeat current predictive control with a high-order terminal sliding mode observer. This high-order terminal sliding mode observer estimates and compensates for the lumped disturbances of the motor model, eliminating steady-state current errors and improving the current's dynamic performance.

[0005] To achieve the above object, the present invention provides the following solutions:

[0006] A deadbeat current predictive control method based on a high-order terminal sliding mode observer includes a permanent magnet synchronous motor model with a lumped disturbance and a high-order terminal sliding mode observer. This method introduces a lumped disturbance into the traditional permanent magnet synchronous motor model to improve model accuracy. The terminal sliding mode observer estimates and compensates for the lumped disturbance, improving the steady-state accuracy and dynamic response of the motor current control.

[0007] A permanent magnet synchronous motor model with lumped disturbances is characterized by:

[0008] The specific expression of the permanent magnet synchronous motor model with lumped disturbance is:

[0009]

[0010] Where R, L, ψ f are the actual values of the stator resistance, inductance and permanent magnet flux of the motor respectively; R0, L0, ψ f0 are the nominal values of the stator resistance, inductance and permanent magnet flux of the motor respectively; F d and F q It is expressed as a lumped perturbation in the dq coordinate system, which includes the error caused by parameter mismatch and the unmodeled dynamic ε d and ε q .

[0011] The expression of the high-order terminal sliding mode observer is

[0012]

[0013]

[0014] Where e d and e q are the current observation errors of dq axis, s d and s q are the sliding surfaces of the dq axes, T is the period of the current loop, k d1 、k d2 、k d3 is the sliding mode gain coefficient of the d-axis current, k q1 、k q2 、k q3 is the sliding mode gain coefficient of the q-axis current, and the k in the bracket represents the state quantity of the k-th current cycle.

[0015] The sliding mode gain in the high-order terminal sliding mode observer must satisfy the relationship

[0016]

[0017]

[0018] Compared with the prior art, the present invention has the following advantages:

[0019] By combining a traditional deadbeat current predictive control structure with a high-order terminal sliding-mode observer, the designed high-order terminal sliding-mode observer can estimate and compensate for the model's lumped disturbances, improving the control system's resistance to disturbances and enhancing the robustness of the current loop. Furthermore, the high-order terminal sliding-mode observer overcomes the oscillation and slow convergence issues of traditional sliding-mode observers, thereby extending the control system's stability range. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required.

[0021] Figure 1 This is a control block diagram of deadbeat current predictive control based on high-order terminal sliding mode observer;

[0022] Figure 2 This is a flow chart of deadbeat current predictive control based on high-order terminal sliding mode observer;

[0023] Figure 3 The current response waveform of the permanent magnet synchronous motor when the nominal parameters are smaller than the actual parameters and there is an error, Figure 3 (a) is the current response when using traditional deadbeat current prediction control. Figure 3 (b) Current response when deadbeat current predictive control is used with a high-order terminal sliding mode observer;

[0024] Figure 4 The current response waveform of the permanent magnet synchronous motor when the nominal parameters are greater than the actual parameters and there is an error, Figure 4 (a) is the current response when using traditional deadbeat current prediction control. Figure 4 (b) Current response when deadbeat current predictive control is used with a high-order terminal sliding mode observer; DETAILED DESCRIPTION

[0025] The present invention will be further described below with reference to the accompanying drawings and examples:

[0026] It should be noted that the following detailed description is intended to further illustrate the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood in the technical field to which this application belongs.

[0027] The purpose of this invention is to provide a deadbeat current prediction control method based on a high-order terminal sliding mode observer. Figure 2 The digital implementation process of this method can be summarized into three steps. Assume that the control cycle of the microprocessor is T, and the dq axis current at time t = k is recorded as i d (k) and i q (k), the dq axis current observation value at time t = k is recorded as and The motor angular velocity is measured and calculated by the encoder and recorded as ω e (k) and the reference current output by the outer loop speed controller is recorded as i d * (k+1) and i q * (k+1), in the kth control cycle, the microprocessor performs the digital implementation of the algorithm according to the following steps within an interrupt.

[0028] Construction of high-order terminal sliding surface: The sliding surface is discretized using the forward Euler method to obtain the discrete equation of the sliding surface.

[0029]

[0030] Current prediction and lumped disturbance calculation: The high-order terminal sliding mode observer is discretized using the forward Euler method to obtain the discrete equation of the observer.

[0031]

[0032] Generation of voltage given signal: Substituting the current prediction value and lumped disturbance into the permanent magnet synchronous motor voltage equation can obtain the discrete equation of the voltage given signal.

[0033]

[0034]

[0035]

[0036]

[0037] In summary, the three steps above fully describe the workflow of a deadbeat current predictive controller based on a high-order terminal sliding mode observer. After the previous control cycle ends, the controller returns to the first step and begins a new cycle.

[0038] In this example, the motor speed is 1000 rpm and the load torque is 1 N·m. Figure 3 (a) and Figure 3 (b) shows the current responses of the traditional deadbeat current prediction control and the deadbeat current prediction control based on the high-order terminal sliding mode observer when the model nominal parameters are smaller than the actual parameters. The actual q-axis current of the traditional deadbeat current prediction control is smaller than the given current, while the actual q-axis current is equal to the given current after the control method in this paper is adopted. The amplitude of the motor change after loading is reduced and the recovery time is shortened. Figure 4 (a) and Figure 4 Figure (b) shows the current responses for traditional deadbeat current predictive control and deadbeat current predictive control based on a high-order terminal sliding mode observer, respectively, when the model nominal parameters are greater than the actual parameters. With traditional deadbeat current predictive control, the actual q-axis current is greater than the set current. However, with the control method proposed in this paper, the actual q-axis current is equal to the set current, and the current ripple amplitude is reduced. Therefore, current predictive control based on a high-order terminal sliding mode observer can eliminate current static errors, improving the dynamic performance and steady-state accuracy of the motor.

[0039] The examples used in this article illustrate the principles and implementation steps of the present invention, and the above examples are used to help understand the method of the present invention and its core ideas.

Claims

1. A deadbeat current prediction control method that introduces a high-order terminal sliding mode observer, comprising a high-order terminal sliding mode observer and a deadbeat current prediction controller. It is characterized in that: A high-order terminal sliding mode observer is introduced into the structure of the deadbeat current predictive controller, wherein the high-order terminal sliding mode observer estimates the motor parameter perturbation and the unmodeled disturbance term, and the deadbeat current predictive controller calculates the reference voltage according to the current given and the estimated value of the disturbance term.

2. The high-order terminal sliding mode observer according to claim 1, characterized in that include: The high-order terminal sliding mode observer sets the disturbance term as the effect of motor parameter perturbation and unmodeled disturbance, which is in the form of Where F d and F q are the lumped perturbations in the dq coordinate system, R, L, ψ f are the actual values of the stator resistance, inductance and permanent magnet flux of the motor, R0, L0, ψ f0 are the nominal values of the stator resistance, inductance and permanent magnet flux of the motor, ω e is the electrical angular velocity of the motor, i d 、i q are the direct-axis and quadrature-axis currents, ε d and ε q represent the unmodeled disturbances in the dq coordinate system respectively. The expression of the high-order terminal sliding mode observer is: Where e d and e q are the current observation errors of dq axis, s d and s q are the sliding surfaces of the dq axes, T is the period of the current loop, k d1 、k d2 、k d3 is the sliding mode gain coefficient of the d-axis current, k q1 、k q2 、k q3 is the sliding mode gain coefficient of the q-axis current, and the k in the bracket represents the state quantity of the k-th current cycle.

3. The deadbeat current prediction controller according to claim 1, wherein: include: The specific expression of the deadbeat current prediction controller is: Where u d and u q are the voltage reference values of the dq axis, i d * and i q * are the current given values of dq axis respectively.

Images