Improvement of direct torque control by using double-fuzzy logic technique

A novel direct torque control strategy, using genetic algorithm on-line to optimize the fuzzy PI controller, is proposed. In this method, according to speed error and its first time derivative, the proportional coefficient K_P and integral coefficient KI can be on-line adjusted by fuzzy adaptive PI speed regulation, and the fuzzy logic adapter parameters are optimized by genetic algorithm to improve the self-adaptation of speed. Moreover, the second fuzzy logic controller is applied to select the voltage vector instead of the conventional hysteresis controllers. In this paper, a comparative study between fuzzy PI direct torque control and proposed approach shows that not only the speed response, overshoot and speed steady precision have been improved, but also the torque, flux and stator current ripples have been effectively decreased at low speed. Simulation results show that the double-fuzzy DTC is more effective than the conventional DTC and that it reduces the torque ripples, and the robustness of the whole system has been enhanced.