Abstract
With the deepening of smart grid construction, the complexity of power grid material warehousing and emergency distribution continues to increase, which puts higher demands on efficient and scalable optimization models. In response to the problems of insufficient search efficiency and weak adaptability to dynamic disaster scenarios in traditional methods, this study proposes an optimization model that integrates a multi-strategy collaborative adaptive chaotic particle swarm optimization algorithm and an improved imperialist competitive algorithm. The experiment is conducted based on multiple types of emergency scenarios, and the results show that the accuracy of material classification of the model is as high as 98.73%, and the delivery time is shortened from 6.78 h to 4.56 h in earthquake scenarios. In the solution of the supply chain distribution path, the single iteration calculation is only 0.45s, which combines stability and efficiency. This model improves the search diversity of inventory parameter optimization by introducing multi-strategy chaotic disturbances and adaptive inertia weights, and combines fishbone warehouse layout to enhance picking efficiency. In the path planning stage, research is conducted to enhance the adaptability of algorithms to complex constraints and dynamic environments through immune penalty correction mechanisms and pheromone adaptive balancing strategies. Research results denote that this model has significant advantages in multi-objective optimization, warehouse layout planning, and emergency logistics path scheduling. This study provides feasible technical solutions for smart grid material management and emergency distribution, as well as new methodological references for complex supply chain optimization research.