Improving the efficiency of induction motor drive by flux and torque control: A hybrid LSE-RERNN approach

A hybrid technique is proposed in this manuscript for the optimal design of an induction motor (IM) drive for the dynamic load profiles during torque and flux control. The proposed hybrid method combines a Ladder-Spherical-Evolution-Search-Algorithm (LSE) and a recalling-enhanced recurrent-neural network (RERNN), which is called an LSE-RERNN technique. The major objective of the proposed method is to minimize IM losses while maintaining control over speed and torque. The proposed method effectively tunes the gain parameter of the PI controller for flux and torque regulation. The LSE methodgenerates a set of gain parameters optimally predicted by RERNN. The method reduces losses without prior knowledge of load profiles, achieving energy savings for steady-state optimum flux. The performance of the proposed technique is done in the MATLAB and is compared with different existing techniques. The value of the proposed method for the mean is 0.328, the standard deviation (SD) is 0.00334, and the median is 0.4173. The loss of the proposed method is much less than 0.3 W while compared to different existing approaches. Moreover, the computation time of the proposed approach is lesser than the existing techniques.