Dynamics of Multi-Objective Energy Optimization in Multi-Energy Micro-Grids Based on Renewable Energy

This paper focuses on the issues of energy sizing and multi-criteria optimization in small and medium power decentralized micro-grids. The application sector concerns water pumping and desalination systems. Two sizing methodologies by optimization are presented to decrease the computation cost in the optimization process. A two-objective optimization is developed based on genetic algorithm: 1) LPSP as technical indicator and 2) embodied energy as environmental indicator. Optimization methodologies are investigated, on the one hand, by a dynamic simulator and a parametric sensitivity study and, in other hand, by a meta-model based on the design of experiment approach. The results show that the process time of the optimization based on meta-model is 16 minutes, which means a decreases of 360 compared to process time of the optimization via dynamic micro-grid simulator. The methodologies and optimization algorithms developed make it possible to provide innovative solutions to the optimization design, especially applied to the electrical energy micro-grid.