Direct Torque Control for Induction Motor Based on Fuzzy-Neural Network Space Vector Modulation

Abstract

AbstractIn order to improve direct torque control (DTC) system dynamic performance and low-speed performance for space vector modulation (SVM) were analysed. The two PI controllers which were used to generate reference voltage vector in conventional SVM–DTC were analyzed. The parameters of PI controller are difficult to determine. A novel control strategy of DTC induction motor based on fuzzy-neural was proposed. The design process of the neural network controller which generates the flux reference voltage vector and fuzzy controller by applying the torque reference voltage vector was represented. Simulations and experiments were carried out to verify the proposed strategy, and the results were compared with conventional SVM–DTC. The simulation and experiment results verify whether the fuzzy-neural network SVM–DTC is capable of effectively improving the control performance, especially improving SVM–DTC system's low-speed performance.KeywordsInduction motorsDirect torque controlSVMFuzzy controlNeural-network controlLow-speed performance