Real-time adaptive speed control of vector-controlled induction motor drive based on online-trained Type-2 Fuzzy Neural Network Controller

Background The speed control of induction motor having superior features such as high efficiency, robust construction, low maintenance and low cost is carried out more effectively with the developments in control theory. With innovations in power electronics and microprocessor technology, it has been made possible to use the vector control method for applications requiring high performance in induction motor drives. Aim and Objective In this study, Type-2 Fuzzy Neural Network (T2FNN) controller which is durable, adaptable and has fast dynamic response capabilities against parameter changes, is proposed to obtain a robust speed response from induction motor. Materials and Methods Matlab/RTI model is developed through DS1103 controller card to experimentally test the speed performance of the proposed controller based induction motor. The proposed controller is trained on-line to improve the robustness of the induction motor against disturbances. After that, experimental studies are built to investigate the speed control behavior and effectiveness of the induction motor. Results and Discussion The performance of T2FNN controller is compared with PI and Type-1 Fuzzy Neural Network (T1FNN) controllers. The experimental results clearly indicate that the proposed controller has a faster and more stable dynamic response capability. Conclusion The proposed controller is significantly improved the dynamic response of the induction motor compared to T1FNN and PI controllers. In addition, the settling time, overshoot and recovery time promoting percentages of T1FNN and PI controllers by T2FNN controller are in satisfactory levels during all steady-state and the transient conditions.