Machine Learning Approach For System Reconfiguration Considering Profit-Driven Security Market

The reconfigurable smart microgrids (RSMGs) represent the novel form for Microgrid (MG) that is worthy of further research. The present study examines a daily risk-based optimal scheduling of RSMG using a wind turbine in order to maximize the profit of the MG operator. A simulation on the basis of the Autoregressive Moving Average model is performed by considering the wind speed, selling energy cost, and buying energy cost as uncertain parameters. The grey wolf optimization algorithm has been developed to determine the optimal combination of MG switches every hour. Additionally, the condition value-at-risk has been employed to formulate a risk measure. According to the simulation outcomes, by evaluating the risk, the predicted gain for the optimum planning issue would increase, and RSMG would be able to generate more revenue through selling energy to the upstream grid over long periods. © 2022