Stochastic Model Predictive Control for Microgrids Based on Monte Carlo Simulations

Distributed renewable generation can be harmonized with the utility grid in flexible structures called microgrids. However, the use of renewables has its drawbacks, such as intermittency and generation uncertainty. Smart controllers can be used to solve such problems and operate the microgrids seamlessly. Accurate forecasts of the generation and demand can be beneficial for optimum operation. Unfortunately, such accurate forecasts may not be available in many cases due to the lack of measurements, the uncertainty of weather conditions, and the human factor. Although renewable sources can be predicted with the state of the art weather forecast methods, there is still uncertainty in their forecasts. Moreover, electric vehicles’ charging time and duration has a probabilistic nature. A stochastic model predictive control methodology is proposed in this work to cope with such scenarios. Throughout the manuscript, the methodology and the corresponding simulation results are presented.