Abstract: The high penetration of renewable energy, heterogeneity of distributed resources, and time-varying nature of loads in microgrids pose severe challenges to their control. Traditional optimization methods often struggle with these nonlinear and non-convex problems, frequently falling into local optima and exhibiting poor adaptability. Metaheuristic optimization algorithms, with their powerful global search capabilities and adaptability to stochastic environments, offer an effective solution for microgrid control optimization. This paper systematically summarizes the applications of metaheuristic algorithms in microgrid control, specifically reviewing their implementation and effectiveness in key areas such as energy management, economic dispatch, resilience enhancement, and anomaly detection. Furthermore, it analyzes current technical challenges in the field and explores future research directions from perspectives of data preprocessing, multi-algorithm fusion, and distributed computing architecture, aiming to provide concrete theoretical reference and practical guidance for the efficient and reliable operation of microgrids.