Stochastic Microgrid Control Problems: Effects of Load Distribution and Planning Horizon

Microgrids enable the integration of distributed energy resources with high renewable penetration into the main power grid. In this study, a microgrid problem that takes into account the stochastic nature of the net load, defined as the difference between actual demand and renewable generation, is studied. The problem is formulated as a Mixed Integer Linear Stochastic Optimization Programming, and is solved under different net load distributions and planning horizons. Numerical results show that increasing variance causes a rise in total system cost, for the approach that solves the stochastic problem by ignoring randomness (as in most real-life applications) as well as for the one that solves the problem with the true distribution. It is observed that enlarging the planning horizon also has similar effects.