Intelligent recursive nonsingular terminal sliding mode control of permanent magnet linear synchronous motor

In order to solve the problem that permanent magnet linear synchronous motor (PMLSM) is easily affected by the parameter variations, external disturbances, friction and other uncertain factors during operation, a recursive nonsingular terminal sliding mode control (RNTSMC) method based on double-hidden-layer radial basis function neural network (DRBFNN) is proposed to improve the control performance of PMLSM system. Firstly, the nonsingular terminal sliding surface and the recursive integral terminal sliding surface are constructed respectively to make the two sliding surfaces arrive successively, which can weaken the chattering and ensure that the tracking error converges to zero in theoretical finite time. However, it is difficult to determine the boundary of the system uncertainties, so the DRBFNN with higher fitting accuracy and generalization ability is introduced to approximate and compensate the uncertainties, and the online adaptive updating is used to weight to further improve the approximation ability of neural network. Finally, the system experiment results show that the method can suppress the influence of uncertainties on the system, which effectively improve the position tracking accuracy of the system and make the system have strong robustness. © 2022 South China University of Technology. All rights reserved.