An efficient MABC-ANN technique for optimal management and system modeling of micro grid

Abstract

In this paper, a hybrid technique is proposed for the system modeling and energy management of Microgrid (MG) connected sources. The proposed hybrid technique is the hybridization of Artificial Bees Colony (ABC) with Bat Search Algorithm (BAT) and Artificial Neural Network (ANN). Here, the novelty of the proposed method is that the search space of the scout bees is gained with the utilization of the BAT and the optimal solution is found from this scout bees search space. The performance of the scout bees is generously enhanced by BAT. So the combination of ABC and BAT is called as MABC technique. The proposed algorithm optimizes the configuration of the MG sources combination by considering the multi-objective function, while fulfilling the load demand. The considered MG connected system such as Wind Turbine, Photovoltaic array, Fuel Cell, Micro Turbine, Diesel Generator and battery storage. Here, ANN is utilized for the prediction of the required load demand by using the inputs of MG connected sources. Based on the load demand, MABC technique is utilized to choose an optimal configuration of MG, i.e., fuel cost minimization, emission factors, operating, and maintenance cost. The proposed technique is executed in MATLAB/Simulink platform and compared with existing strategies. In this paper, the cost accuracy percentage (CAP) for proposed and the existing technique is also investigated, the proposed technique achieves 7.37880 %.