Cost Sensitivity Analysis to Uncertainty in Demand and Renewable Energy Sources Forecasts

Addressing uncertainty has become a necessity when modeling modern power systems. Many state-of-the-art methods suffer from either poor uncertainty characterization or a high computational burden. This paper proposes a model that is easy to implement, fast to compute, and effective in addressing uncertainty. It is based on the model predictive control algorithm with the addition of uncertainty parameters optimization. For demonstration purposes, the model is applied to a microgrid consisting of a wind turbine, a local load, and battery energy storage. The model seeks to satisfy the local demand at the lowest cost by procuring energy from the battery energy storage, the wind turbine (in its portfolio), or the wholesale market, where wind power output, local demand, and market prices are uncertain parameters. In the presented case study, the upper bounds obtained using our model are close to the perfect information deterministic model values. Hence, this model has a great potential for practical use.