Instantaneous torque control of translational meshing motors based on neural network

Abstract

An adaptive neural network control of a novel type of translational meshing motor with model uncertainties is considered. Owing to its nonlinear characteristic, a model reference control system which consists of two neural networks is used. The torque model is identified based on BP neural network, and then a RBF neural network works as the controller. The model reference control system is trained for the optimizing parameters and research of control strategy. The description of the control system and training procedure of the two neural networks are given. The test results obtained for a torque control scheme suitable for the control of the motor are also presented to verify the effectiveness of the proposed nonlinear control scheme. It has been found that the neural network control system is able to work reliably.