Energy Management of Virtual Power Plant Using Stochastic Programming Approach

In this paper, a stochastic programming model is proposed for energy management of a Virtual Power Plant (VPP) including solar power stations, energy storage facilities and demands interconnected within a microgrid. In the stochastic programming model, uncertainty in availability of microgrid components is modeled as a scenario set. Additionally, uncertainty in energy prices and solar power productions is modeled as single-point prediction. Energy management of the VPP is implemented in two stages of decision-making and power-dispatch in each time period. An energy management system (EMS) determines the power traded with main grid in decision-making stage considering probable contingency scenarios. In power-dispatch stage, the EMS makes appropriate decisions on energy consumption and production, while monitoring conditions of microgrid components and updating available information in real-time. The energy price and solar energy production are forecasted based on real-world historical data collected from the New England electricity market, US. Results obtained by a case study indicate performance of the stochastic programming model compared to robust optimization and deterministic models.