An optimal control for power management in super capacitors/battery of electric vehicles using Deep Neural Network

Abstract

Nowadays, Hybrid Energy Storage Systems (HESS) is gaining popularity as a result of their superior system efficiency and battery lifetime when compared to solitary energy sources. To maximize the energy management for electric vehicles, HESS like batteries and super capacitors (SCAP) are used, which has two objectives: (i) first the voltage of SCAP reference can be determined viaincluding real-time dynamics of load and (ii) optimize the power flow by reducing the magnitude variation of battery power & power loss. A sophisticated model of the DC-DC converter is taken into account in this HESS power management issue, which includes both conduction and switching losses. Also, the optimization issue is then quantitatively tackled for varied drive cycles utilizing a Proportional Integral Derivative(PID) controller.Therefore, to train and predict the control parameters of HESS, a Deep Neural Network (DNN) is used that consists of layers of neurons between the input and output layers, which fuse the feature extraction process with an increase in accuracy. Hence, to generate the optimal controllerparameters of HESS, it's planned to propose a novel meta-heuristic algorithm called the Squirrel Search with Improved Food Storage (SS-IFS) Optimizer, which is the conceptual improvement of the standard Squirrel Search Algorithm (SSA).Finally, the superiority of the proposed approach is demonstrated using various metrics.