ANN-Based Binary Backtracking Search Algorithm for VPP Optimal Scheduling and Cost-Effective Evaluation

Abstract

This article reports an artificial neural network (ANN)-based binary backtracking search algorithm (BBSA) for optimal scheduling controller applied on IEEE 14-bus system for controlling microgrids (MGs) formed virtual power plant (VPP) toward sustainable renewable energy sources (RESs) integration. The model of VPP was simulated and validated based on actual parameters and load data reported in Perlis, Malaysia. BBSA optimization algorithm offers the best binary fitness function to find the best cell. It creates the optimum scheduling using the actual data for wind speed, solar radiation, fuel conditions, battery charging/discharging, and specific hour demand. The developed ANN-based BBSA search for the optimal ANN parameters architecture, e.g., (the number of neurons and learning rate) that enhanced the ANN controller to predict the optimal schedules to regulate power-sharing via prioritizing the utilization of RES in place of the national grid purchases. The results of the optimal <sc xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">on/off</small> status prediction of the 25 DGs showed that the ANN-BBSA gives a mean absolute error (MAE) of 6.2 × 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">−3</sup> with a unity correlation coefficient. The results showed a significant reduction in the cost and emission by 41.88% and 40.7%, respectively. Thus, the developed algorithms reduced the energy cost while delivered reliable power toward grid decarbonization.