An implementation of soft computing approach of smart control for induction motor using ANFIS
S. Sujitha, K. Vinoth Kumar, R V Shiva, Sagar Kulkarni, M M Ponnappa
Abstract
This paper presents a comparative analysis of the performance of a 3-Phase Induction motor by using Adaptive Neuro Fuzzy Inference System (ANFIS or Neuro-fuzzy) and Proportional-Integral-Derivative (PID) controllers. This study is mainly to show improved speed control of a scalar closed loop Induction motor with an ANFIS controller which also helps to keep the motor speed constant when the load varies. This comparative analysis done on some of the main properties on which the efficiency of an Induction motor depends on. DCT or Direct torque control scheme is one of major advanced methods to control the electromagnetic torque and flux of machines. In these drives the control of torque and speed for high performance applications demands a high robust adaptive controller. This is where ANFIS comes in which is a combination or a hybrid of ANN (Artificial Neural Network) and FLCC (Fuzzy Logic Control) which helps in the execution of direct torque control and overcome the difficulties in high performance machines ad drives. Using the fuzzy toolbox in MATLAB and SIMULINK, the graphical results of both the controllers are determined and why Neuro-fuzzy controller is a better option and can be used in Real-time comfortably is explained. The Neuro-fuzzy controller also shows enhanced performance and dynamics of the induction motor when compared to a normal PID controller.