Energy management system using artificial fish swarm speed optimized fuzzy controller based on a deep recurrent neural learning classifie

Abstract

Hybrid Electric Vehicles (HEVs) must ensure power demand through minimum fuel consumption and a control strategy. Existing control methods were easy to implement, showing quick response and good performance. Power demand is linked to numerous factors such as level of social and economic expansion, industrialization, urbanization, and technological growth. However, power demand problems like higher energy waste, poor quality, less accuracy, lack of robustness, and limited operating range were not reduced in existing controller  methods. This paper presents an Artificial Fish Swarm Speed Optimization Fuzzy PID Controller (AFSSOF-PIDC). AFSSOFPIDC-DRNLC includes different layers in drive train management. Initially, different vehicle data is considered in the input layer and then sent to hidden layer 1. Fitness is identified by improved Artificial Fish Swarm Speed Optimization to find optimal values that minimize the power demand, and then send it toward hidden layer 2. A Mamdani Fuzzy PID Controller is used in hidden layer 2. If the fitness value of the vehicle information is less than the threshold value, fuel consumption is minimized in the HEV. Otherwise, consumption of fuel is not minimized in the HEV. Finally, energy management is achieved through minimal power demand. The results indicate that the performance of the proposed AFSSOFPIDC-DRNLC technique minimizes fuel consumption by increasing the performance of the controller as compared with existing methods.