Abstract: With increasing global concern over climate change and fossil fuel depletion, the proportion of renewable energy in power systems continues to rise. However, the randomness and volatility of wind and photovoltaic power generation pose challenges to grid dispatching. The introduction of energy storage systems has emerged as an effective means to balance supply and demand while improving energy utilization efficiency. This paper proposes an optimized scheduling method based on an improved Northern Goshawk Optimization for multi-energy microgrid systems. The approach comprehensively considers various energy forms, including wind power, photovoltaic power, fuel cells, micro gas turbines, and energy storage batteries, aiming to minimize the comprehensive daily operational costs of the microgrid. It constructs an integrated optimization model encompassing generation costs, maintenance costs, interaction costs, and environmental costs. To enhance the global search capability and convergence performance of the algorithm, the traditional Northern Goshawk Optimization incorporates a reverse learning mechanism, a sine-cosine dual string search strategy, and a cubic chaotic mapping operator to reduce system operational costs, improve renewable energy utilization, and enhance the economic efficiency and stability of the microgrid.