An AI-Based Algorithm for Energy Management of Hybrid Renewable Energy System

In recent times energy demand has increased due to several factors, such as the growing world population, conventional ways of energy transmission & distribution, adopting less energy-efficient technologies, etc. These factors contribute towards more utilization of natural resources to meet the energy requirements, thus polluting the environment by producing harmful gases and increasing the energy cost. This paper addresses the aforementioned challenges, where it aims to look for more sustainable ways of energy production such as solar energy, wind energy, and biomass. This paper designed an off-grid system that aims to meet the University of Palermo's energy demand at a minimum annualized system cost. In this design, two ways are adopted for energy storage: 1) battery storage, and 2) hydrogen-based storage. The priority is to charge the batteries when excess energy is available. When the battery is fully charged and still some energy is available, this energy is used to operate the electrolyzer to produce hydrogen, which can also be used to produce again energy through fuel cells. This paper implemented an improved grey wolf optimization algorithm to find optimal power flow. The advantage of using it over other optimization algorithms is that it reduces premature convergence, and finds the global best solution. The result shows that, when the sources are not enough, the storage system can release enough energy for all time in the day.