Multi-objective Optimization in Self-driving System of Distribution Network: Adopting PSO Strategy

In the existing literature, a variety of distribution network optimization methods have been proposed, including genetic algorithm, ant colony algorithm and simulated annealing algorithm. However, these methods have many optimization variables and complex constraints, and the traditional optimization methods are easy to fall into local optimal in the solving process. The main objective of this paper is to design a multi-objective optimization algorithm based on PSO for the optimal scheduling of autonomous vehicle systems in distribution networks. Particle swarm optimization (PSO), as a new swarm intelligence optimization algorithm, has become an effective tool to solve multi-objective optimization problems because of its simplicity, global search ability and fast convergence speed. The application of PSO strategy to the multi-objective optimization of distribution network autonomous vehicle system is expected to overcome the shortcomings of traditional optimization methods and achieve more efficient and reliable optimization results. The main contribution of this paper is to effectively balance multiple optimization objectives and improve the overall performance of the system. Under single objective optimization, the average driving distance from vehicle 1-5 to area 1-5 was above 3 kilometers; under multi-objective optimization, the average driving distance of vehicles 6-10 to areas 1-5 was mostly within 3 kilometers. By optimizing the operation and management of the distribution system, sustainable development of the power system can be achieved, reducing its impact on the environment and improving social and economic benefits.