A retired power battery recycling network optimization model and intelligent algorithm

With the rapid growth of electric vehicle sales, the research on the recycling network of retired power batteries urgently needs to be developed synchronously, whose common deficiencies include insufficient consideration of network layer complexity and too single optimization objective. So, a retired power battery recycling network optimization model and intelligent algorithm are proposed to achieve multi-objective and multi-level integrated optimization, which could implement the extended producer responsibility system and appropriately utilize the existing forward logistics network. Firstly, considering both the economic costs and environmental costs, a retired power battery recycling network optimization model based on bi-level programming is proposed to achieve integrated optimization at the strategic and tactical levels, which takes both the minimum total cost of retired power batteries reverse logistics and the minimum negative impact on residents as the upper-level objective and takes the shortest tour path of recycling vehicles as the lower-level objective. Secondly, insufficient diversity of solution sets and slow convergence speed are usually shortcomings of existing heuristic algorithms in solving such planning models, so a multi-objective grey wolf and genetic algorithm (MGW-GA) is proposed to solve the model by using leader selection strategies and external storage archives and applying the ideas of evolutionary game and hierarchical iteration. Finally, the feasibility and effectiveness of the constructed model are proved by experiments. The performance of algorithm is compared with an existing excellent heuristic algorithm, NSGA-II-GA. The results indicate that the MGW-GA exhibits significant advantages in terms of the diversity and uniformity of the non-dominated solution set as well as convergence speed.