RBF Neural Network Application in Internal Model Control of Permanent Magnet Synchronous Motor

2011 ◽  
pp. 68-76 ◽  
Author(s):  
Guohai Liu ◽  
Lingling Chen ◽  
Beibei Dong ◽  
Wenxiang Zhao
Keyword(s):  
Neural Network ◽  
Permanent Magnet ◽  
Internal Model ◽  
Rbf Neural Network ◽  
Permanent Magnet Synchronous Motor ◽  
Synchronous Motor ◽  
Internal Model Control ◽  
Model Control ◽  
Network Application

scholarly journals Current Control Methods for an Asymmetric Six-Phase Permanent Magnet Synchronous Motor

Electronics ◽  
2020 ◽  
Vol 9 (1) ◽  
pp. 172 ◽  
Author(s):  
Zhihong Wu ◽  
Weisong Gu ◽  
Yuan Zhu ◽  
Ke Lu
Keyword(s):  
Permanent Magnet ◽  
Internal Model ◽  
Control Method ◽  
Permanent Magnet Synchronous Motor ◽  
Synchronous Motor ◽  
Current Control ◽  
Internal Model Control ◽  
Lms Algorithm ◽  
Least Mean Square ◽  
Model Control

Via the vector space decomposition (VSD) transformation, the currents in an asymmetric six-phase permanent magnet synchronous motor (ASP_PMSM) can be decoupled into three orthogonal subspaces. Control of α–β currents in α–β subspace is important for torque regulation, while control of x-y currents in x-y subspace can suppress the harmonics due to the dead time of converters and other nonlinear factors. The zero-sequence components in O1-O2 subspace are 0 due to isolated neutral points. In α–β subspace, a state observer is constructed by introducing the error variable between the real current and the internal model current based on the internal model control method, which can improve the current control performance compared to the traditional internal model control method. In x–y subspace, in order to suppress the current harmonics, an adaptive-linear-neuron (ADALINE)-based control algorithm is employed to generate the compensation voltage, which is self-tuned by minimizing the estimated current distortion through the least mean square (LMS) algorithm. The modulation technique to implement the four-dimensional current control based on the three-phase SVPWM is given. The experimental results validate the robustness and effectiveness of the proposed control method.

Position control for permanent magnet synchronous motor based on neural network and terminal sliding mode control

2020 ◽  
Vol 42 (9) ◽  
pp. 1632-1640
Author(s):  
Wenwu Zhu ◽  
Dongbo Chen ◽  
Haibo Du ◽  
Xiangyu Wang
Keyword(s):  
Neural Network ◽  
Permanent Magnet ◽  
Finite Time ◽  
Position Control ◽  
Control Method ◽  
Sliding Mode ◽  
Rbf Neural Network ◽  
Permanent Magnet Synchronous Motor ◽  
Synchronous Motor ◽  
Terminal Sliding Mode

A finite-time control strategy is proposed to solve the problem of position tracking control for a permanent magnet synchronous motor servo system. It can guarantee that the motor’s desired position can be tracked in a finite time. Firstly, for the d-axis voltage, a first-order finite-time controller is designed based on the vector control principle. Then, for the q-axis voltage, based on a radial basis function (RBF) neural network, an integral high-order terminal sliding mode controller is designed. Theoretical analysis shows that under the proposed controller, the desired position can be tracked by the motor position in a finite time. Simulation results are given to show that the proposed control method has a strong anti-disturbance ability and a fast convergence performance.

Sign in / Sign up

Export Citation Format

Share Document