Powell Metaheuristic Cat Swarm optimized Sugeno Fuzzy Controller based Deep Belief Network for energy management in Hybrid electric vehicles

Abstract

Hybrid electric vehicles (HEV) have joined the Internal Combustion Engine (ICE) scheme using electric propulsion system, i.e. HEV technology combines the mechanical drive train and electric vehicle. Hybrid vehicles need the supervisory algorithm with energy management strategy to drive the electric vehicles. An energy management aim is used for reducing fuel towards constraints, vehicle as well as driver. In order to implement energy management techniques, Powell Metaheuristic Cat Swarm Optimized Sugeno Fuzzy Controller based Deep Belief Network (PMCSOSFC-DBN) Model is introduced for energy management in hybrid electric vehicle with minimal fuel consumption. PMCSOSFC-DBN Model comprises various layers for minimizing fuel consumption. Initially, the vehicle speed, engine speed, motor speed and state of charging are taken by input and sent towards input layer. Input layer sends the collected information to the hidden layer 1 where Powell Metaheuristic Cat Swarm Optimization is performed for determining fitness function to identify optimal values. After that, fitness value of vehicle information is sent to hidden layer 2 which Sugeno Fuzzy PI Controller is employed in hybrid electric vehicles to manage vehicle speed for minimizing the fuel consumption. By this way, an energy management with lesser power demand is carried out within HEV. PMCSOSFC-DBN is determined in fuel, vehicle speed as well as SOC. The simulation results show that PMCSOSFC-DBN minimizes the fuel consumption for enhancing the controller performance than the conventional methods.

URN:NBN:sciencein.jist.2023.v11.488