The management and scheduling of DGs, including renewable energy sources, has been the subject of much research
[2][3][4]. For example, in
[4] the scheduling and operation of islanded multi-microgrids using decentralized collaborative dispatch framework and multi-agent consensus algorithms were investigated. Regarding the inherent uncertainty of renewable energy sources, the problem of predicting wind speed and solar radiation has been studied in
[5][6][7][8]. In
[5], the evolutionary optimized local general regression neural network was used in order to predict wind speed and solar radiation. In
[6], a stochastic scheduling model was proposed for CHP-based microgrids, by considering a periodic or seasonal pattern of load and price processes in the scenario generation procedure by ARIMA models. In
[7], an adaptive modified firefly algorithm was used to solve the renewable MG, by considering the uncertainty of load forecast error, renewable energy sources, and market price. In
[8], the intraday rolling dispatch strategy for the off-grid CHP microgrid was proposed to overcome renewable energy sources’ uncertainties. In
[9], an intelligent microgrid energy management methodology is proposed by using fuzzy environment and consideration of the uncertainty of renewable energy sources, in order to minimize power losses and costs. In
[10], a stochastic scenario-based model for uncertainty modeling of demand, power generation of wind turbines, solar cells, and electricity market prices was presented. In
[11], the optimal microgrid operation including fuel cells and CHP units was studied via the particle swarm optimization (PSO) algorithm; it also examined the impact of different tariffs on electricity sales and purchases every hour of the day. In
[12], a mathematical scheduling for the optimal operation of micro-CHP units in the microgrid was presented.