A stochastic MPC based approach to integrated energy management in microgrids

In this paper, a stochastic model predictive control (SMPC) approach to integrated energy (load and generation) management is proposed for a microgrid with the penetration of renewable energy sources (RES). The considered microgrid consists of RES, controllable generators (CGs), energy storages and various loads (e.g., curtailable loads, shiftable loads). Firstly, the forecasting uncertainties of load demand, wind and photovoltaic generation in the microgrid as well as the electricity prices are represented by typical scenarios reduced from a large number of primary scenarios via a two-stage scenario reduction technique. Secondly, a finite horizon stochastic mixed integer quadratic programming model is developed to minimize the microgrid operation cost and to reduce the spinning reserve based on the selected typical scenarios. Finally, A SMPC based control framework is proposed to take into account newly updated information to reduce the negative impacts introduced by forecast uncertainties. Through a comprehensive comparison study, simulation results show that our proposed SMPC method outperforms other state of the art approaches that it could achieve the lowest operation cost.