Optimal Sizing Assessment of an Islanded Microgrid Based on Different EMS Strategies

Abstract

Microgrids (MG) optimal sizing considering the minimization of life cycle emissions (LCE) of all MG components as well as their investment and operation costs was a main contribution of our recent studies [1], [2]. However, the design of the energy management system (EMS) in the MG is not the same in these studies. The MILP algorithm defining an EMS with minimal LCE generates power flow distribution on real-time basis in [1] while in [2] the energy management depends on future load power profile and available solar power in order to make the decision. A Genetic Algorithm (GA) is used to minimize in parallel two objectives: LCE and MG costs. The battery usage cost is considered just in the study with non real-time EMS to make sizing more accurate while compromising computation time. The goal of this paper is to compare the optimal sizing of the MG components obtained for a real-time EMS and non real-time EMS in order to evaluate the sizing robustness and accuracy. A new modelling approach was used to allocate LCE and costs on an hourly basis, which allows to avoid specifying when MG elements replacement occurs.