Optimal design for SCAP/battery power management applied in electric vehicle (EV) applications: a KHO-RDF technique

This dissertation proposes power management of optimal control scheme for hybrid energy storage system (HESS) like super capacitor and (SCAP) battery in electric vehicles. The proposed technique is a parallel performance of both the random decision forest (RDF) and krill herd optimization (KHO), and thus, it is called as KHO-RDF method. The main objective is to minimize the difference between the actual and reference power in the battery and SCAP. Here, the HESS framework comprises of two sections: (1) figuring the SCAP reference voltage dependent on load dynamics. (2) Maximizing the power flow through HESS. The reference voltage of SCAP by evaluating real-time load dynamics is computed at first, i.e., the vehicle dynamic, motor characteristics, regenerative braking systems and driving conditions. Furthermore, at the same time the magnitude variety of battery power was minimized and the power loss will occur. The input parameters of SCAP are load current, battery current and state of charge. In proposed technique, possible control signals dataset of HESS is fused to produce KHO. By utilizing the practiced dataset of KHO, the RDF is trained and predicts the optimal parameters of HESS. Moreover, the proposed technique advances the SCAP voltage, battery current magnitude, battery current varieties and battery power. With the proposed approach, the parameter of HESS is optimized and it provides certain solutions. The proposed technique is executed in Matrix Laboratory (MATLAB)/Simulink working platform. By using the comparison analysis with the existing procedures, the performance of the HESS is surveyed.