Forecast-Driven Power Planning Approach for Microgrids Incorporating Smart Loads Using Stochastic Optimization

Abstract

The main focus of this work is on improving the resilience of the standalone microgrids using the available forecast of both energy sources and loads, provided that some smart loads are incorporated in the system. The proposed approach assumes that the smartness of the loads facilitates an accurate prediction of their consumption as well as the possible time span for their operation. The kernel of the proposed approach is to provide a greedy-based scheduling scheme of a group of non-preemptive loads in order to reduce the net deficit between the aggregate load and the low-price power profile, and therefore, the levelized cost of energy (LCoE) can be minimized. Based on the driven forecast, the power routings between the components of the microgrid are investigated considering the scheduling candidates of the incorporated smart loads. Thus, the optimal schedule is selected so as to ensure the maximum utilization of the low-price power. A stochastic optimization method based on the genetic algorithms (GAs) is used to cut-down the massive searching space and provide the optimal schedule within a reasonable time. An illustrative example is used to carry out this work using a group of synthetically created loads representing different facilities inside a hospital in Gaza city. Simulation results show that the proposed algorithm can significantly reduce LCoE and meanwhile maximizing the utilization factor of the installed renewable energy sources.