Accurate modeling of uncertainties based on their dynamics analysis in microgrid planning

Abstract

Abstract The accurate modeling of uncertainties is one important issue in planning of microgrids. The more accurate modeling of the uncertainties results in answers of the stochastic optimization methods which are close to the actual optimal values. This paper utilized a scenario-based stochastic optimization method, considering photovoltaic (PV) and wind turbine power output and load as the uncertain variables in determining the capacity of distributed generation resources in an autonomous microgrid. Costs were considered as the objective function while the grid reliability was the constraint. To generate the scenarios of variables with uncertainty in microgrid, the dynamics of variables was first characterized by the Recurrence Plot nonlinear analysis, followed by the study of their seasonal behaviors. Based on this analysis, a method for classifying data was presented according to their seasonal behavior and the appropriate method to generate the scenarios for each of the uncertain variables was determined. The results showed that generating the scenarios based on the analysis of the uncertain variables results to an increase in the optimization accuracy.