This article provides a comprehensive review of energy management systems (EMS) in microgrids, which are small-scale power grids that can operate independently or connected to the main grid. The primary focus is on managing distributed energy resources (DERs) such as wind turbines and solar photovoltaic modules, which are subject to variability and uncertainty due to their renewable nature. The article highlights the importance of EMS in ensuring stable and reliable energy supply, optimizing resource utilization, and reducing costs and emissions. The review covers various EMS approaches, including classical methods, meta-heuristic approaches, stochastic and robust programming, model predictive control, artificial intelligence (AI), and multi-agent systems. Each approach is evaluated based on its effectiveness, scalability, and adaptability to different microgrid scenarios. The article also discusses the challenges and limitations of these methods, such as computational complexity, data quality, and system stability. Several case studies and applications are presented to illustrate the practical implementation of EMS in microgrids. These include optimizing energy trading, reducing operating costs, and enhancing system reliability through the integration of renewable energy sources. The article concludes by emphasizing the potential of AI technologies in future microgrid operations, while also addressing the challenges and considerations for their broader adoption.This article provides a comprehensive review of energy management systems (EMS) in microgrids, which are small-scale power grids that can operate independently or connected to the main grid. The primary focus is on managing distributed energy resources (DERs) such as wind turbines and solar photovoltaic modules, which are subject to variability and uncertainty due to their renewable nature. The article highlights the importance of EMS in ensuring stable and reliable energy supply, optimizing resource utilization, and reducing costs and emissions. The review covers various EMS approaches, including classical methods, meta-heuristic approaches, stochastic and robust programming, model predictive control, artificial intelligence (AI), and multi-agent systems. Each approach is evaluated based on its effectiveness, scalability, and adaptability to different microgrid scenarios. The article also discusses the challenges and limitations of these methods, such as computational complexity, data quality, and system stability. Several case studies and applications are presented to illustrate the practical implementation of EMS in microgrids. These include optimizing energy trading, reducing operating costs, and enhancing system reliability through the integration of renewable energy sources. The article concludes by emphasizing the potential of AI technologies in future microgrid operations, while also addressing the challenges and considerations for their broader adoption.