Efficiency and Cost Optimization of Three-ϕ Induction Motor Using Soft Computing Techniques

Abstract

In this paper, artificial neural network (ANN) and adaptive neuro-fuzzy inference system (ANFIS) approach are used to evaluate the performance of a vector-controlled induction motor. Adaptive neural-fuzzy inference system (ANFIS) is a mixed system which combines artificial neural network (ANN) and fuzzy logic control (FLC) which improves the performance of the drive. An improvement in efficiency of the motor is obtained by adjusting the flux of the induction motor, and this flux produces the optimal value of current (Ids) which improves the efficiency of induction motor. Firstly, the optimal values of current are obtained by using mathematical expression, and then, these optimal values are incorporated in ANN controller as well as in ANFIS controller. The result obtained demonstrates that ANFIS controller shows better results as compared to ANN controller. A sufficiently great amount of energy can be saved by using ANFIS controller. The whole circuit is simulated using MATLAB 2015a.