Hybrid poisson-lagrange approach based on genetic algorithm for improving performance of induction motor

Abstract

ABSTRACT This article presents an intelligent approach for controlling the mechanical parameters like speed, rotor position, and torque of an induction motor. These parameters have poor performance in terms of settling time, rise time, peak overshoot, and voltage deviation. To conquer such an issue, a hybrid Poisson-Lagrange (HPL) algorithm is proposed. This algorithm is a combination of Poisson distribution and the Lagrange multiplier. It begins with the mathematical modeling of a three-phase induction motor. A multi-objective function is framed with the help of the HPL algorithm. Further, unknown variables of the multi-objective function are trained with a genetic algorithm (GA), which is coined the HPL-GA algorithm. Similar types of control for mechanical parameters are also analyzed with fuzzy logic controllers (FLC) and field-oriented controllers (FOC). It is observed that peak overshoot, rise time, settling time of speed and torque, and sensitivity are found to be minimum with HPL-GA in comparison to FLC and FOC for various loading conditions. The voltage deviation is also found to be lowest with HPL-GA with respect to others for various loading conditions and thus such an approach makes the system highly efficient and robust.