Adaptive neural network control of electric motors in electric inspection vehicles

In this paper, a torque loop control scheme combining adaptive decoupling control (ADEC) and radial basis neural network (RBFNN) is used to improve the effects of parameter variations and load fluctuations of the permanent magnet synchronous motor (PMSM) of the electric inspection vehicle in the grid system. On the one hand, an adaptive decoupling controller is designed based on the mathematical model of the PMSM parameters under external perturbations, and the control strategy is proved to be asymptotically stable according to Lyapunov stability; on the other hand, since the control rate contains the parameters $k_{1}, k_{i d}, k_{2}, k_{i q}$ and affects the dynamic performance of the torque loop control of the PMSM, for this reason, the RBFNN parameters are used for optimization in order to achieve the system’s optimal performance. Finally, the simulation verification shows that the torque and current in the proposed control scheme have strong resistance to fluctuation and fast response speed.