Combined Hcs-Rbfnn For Energy Management Of Multiple Interconnected Microgrids Via Bidirectional Dc-Dc Converters

In this manuscript, a hybrid control method is implemented to model and design bidirectional DC- DC converters for the energy management of interconnected renewable energy sources (RES). The proposed hybrid control method is the joint execution of Hybrid Crow Search Algorithm (HCSA) and Radial Basis Function Neural Network (RBFNN). The modeling and design of the Bidirectional DC-DC converter modeling and design topology is developed with the improved efficiency of the converter, efficient use of renewable energy sources, and the reduction of switching loss. In the proposed manner, the HCSA runs the evaluation procedure for establishing correct control signals for the system and creates the control signal database for offline mode in light of power range among source side (RES, battery and supercapacitor (SCAP)) and load side. Here, the crow's seeking behavior is modified by crossover and mutation. The dataset obtained is utilized to operate the AI approach for the online method and leads the control process on less execution time. In the proposed method, the intention function is defined using the data of the system subject to equality and inequality constraints. The constraint is the availability of RES, power requirement and charge level of storage elements. Batteries and SCAP are utilized as an energy source to allow renewable energy system units to operate continuously in stable and constant power output. At that point, the proposed model is implemented on MATLAB/Simulink work platform and the implementation is assessed to the existing techniques like base method, ALORNN and CSA. Furthermore, switching losses, conduction losses and converter efficiency are also analyzed. Switching losses, conduction losses and converter efficiency of the proposed technique are 0.19W, 0.43W and 98%. (c) 2020 Elsevier B.V. All rights reserved.