In the emergency management of power system, efficient material dispatching is of great significance for rapid response to emergencies. In view of the challenges of traditional scheduling methods in complex multi-objective optimization, this paper explores a multi-objective hierarchical scheduling method based on differential evolution algorithm. The upper layer focuses on the timeliness and economy of resource scheduling, while the lower layer focuses on the fairness and coverage of material allocation. The results show that the optimized algorithm performs well in inverse generation distance, dispersion, scheduling fairness and resource coverage satisfaction, which verifies the adaptability and stability of the algorithm. This method provides a reference for power enterprises to make efficient decisions in emergency management, and helps to reduce the risk of power system and improve the reliability of power supply.