Robust Data Predictive Control Framework for Smart Multi-Microgrid Energy Dispatch Considering Electricity Market Uncertainty

With the emerging technologies for Energy Intent (EI) and data-driven applications, conventional power grid network is undergoing a radical modernization. An efficient energy management electricity price forecasting remains challenging task. In this paper, new Robust Data Predictive Control framework Management System (RDPC-EMS) developed to overcome uncertainty of retail market minimize total operating costs multi-microgrids (MMG) system. The proposed solves economic dispatch based on an accurate Electricity Price Forecasting (EPF) by Outlier-Robust Extreme Learning Machine (OR-ELM) algorithm two layers cooperative Distributed Model (DMPC). First level provides optimal scheduling between Distribution Operator (DSO) microgrids systems cost forecasted price. contrast, second maintains supply-demand balance applying from first layer through adjustment distributed resources (DER). prediction assessed using real dataset Iso New England market. OR-ELM regression method shows significant performance in terms error metrics. For instance, mean absolute training stage 2.05% with comparison 4.17% 6.29% Support Vector Regression (SVR), Artificial Neural Network (ANN) models respectively. Finally, simulation results demonstrate efficiency RDPC-EMS daily reduction, decrease 15% MG 1 16% 2.