Optimization of substation cable laying paths based on the PSO-ACO fusion algorithm

Addressing the issues of local redundant turning points and premature convergence often encountered in traditional ant colony algorithm (ACA) for substation cable laying path planning, an improved strategy for the PSO-ACO fusion algorithm is proposed. A grid method is employed to establish a laying environment model, and a comprehensive fitness function is constructed based on the total cable length and the number of turns. PSO is utilized to pre-search for suboptimal paths, which are then mapped to differentiated initial pheromones for ACO. Additionally, pheromone increments are dynamically allocated between high-quality and ordinary paths. Matlab simulation comparisons demonstrate that the improved algorithm reduces the path length by approximately 11.2% compared to traditional ACO, validating the effectiveness of the fusion algorithm in shortening the path length.