Abstract: This paper proposes an improved Goose Algorithm with adaptive weights for photovoltaic-storage microgrid scheduling. It aims to solve the problems of slow convergence and being easily trapped in local optima in traditional swarm intelligence optimization algorithms. Firstly, this paper builds a scheduling model for a photovoltaic-storage microgrid. The goal is to reduce the total operating cost of the system. The model considers power balance, energy storage charging and discharging, and SOC constraints. Secondly, the improved AWGA is used to balance global search and local optimization, while updating the positions of individuals to improve the convergence performance of the algorithm. Finally, simulation analysis is carried out to compare the improved algorithm with SSA, GWO, and WGA. The results show that the improved algorithm can reduce economic cost and perform well in convergence speed and stability.