Short-term operation of a hybrid minigrid under load and renewable production uncertainty

Properly sized and operated hybrid minigrids can assure good-quality electricity to rural households at an affordable price. A system composed by renewable sources, a diesel generator and batteries can be a cheaper option, but it requires daily operation in order to reduce fuel consumption, to assure fuel availability, and to avoid quick degradation of the batteries. Variability in load and renewable generation introduces uncertainties that must be considered in order to assure a proper operation, thus reducing the curtailment of both load and renewable production. This paper proposes a procedure for short-term operation of a hybrid minigrid in order to cope with errors in forecasting of both load and renewable generation. A probabilistic tool based on Monte Carlo simulations and mixedinteger programming is developed to estimate the optimal working point of the diesel generator and batteries. The Monte Carlo scenarios are singularly optimized, thus defining several optimal schedules that are combined to define the proposed stochastic commitment and dispatchment. The methodology is supported by numerical case studies that even confirm the applicability of Monte Carlo simulations to the short-term operation.