The Optimal Use of Electrical Energy Using Conventional and AI Methods

Abstract

In recent years, there has been a surge of interest among scientists in the field of Microgrid Energy Management. The goal is to find ways to reduce the cost of generating and operating a Microgrid, while maintaining crucial factors such as voltage and frequency stability. This benefits both producers and consumers of electricity. Researchers have taken different approaches to optimize Microgrids and cut down expenses. These include Autonomous building energy management, Control algorithms, Selection of power-generating units, and Scheduling and Planning of Microgrid sub-units. One promising approach is the use of Reinforcement Learning (RL), a type of machine learning, to efficiently schedule and plan storage systems attached to Microgrids. This can help save costs and increase reliance on local renewable energy sources such as Photovoltaic (PV), Wind, and Hydro. In this comprehensive study, we focus on the application of RL for Microgrid optimization and management, with a particular emphasis on scheduling battery storage systems. We also discuss the major challenges, limitations, and future prospects in this area. The benchmark algorithm that compares RL with mixed-integer linear programming (MILP) makes this literature review a novel contribution to the field.