A Novel Co-Evolutionary Multi-Objective Optimization Algorithm

To improve the search efficiency of optimization algorithms and solve issues related to local search, this paper proposes a novel cooperative co-evolutionary multi-objective algorithm. Firstly, the estimation of distribution algorithm is used to accelerate the convergence rate to get the optimal solution, and a "fundamental change" strategy is adopted to improve cooperation between individuals and the evolution of the population, enhancing the global and local search capabilities of the algorithm. Secondly, a straightforward elite-based parent population generation strategy is adopted, which greatly reduces the consumption of computing resources. Through simulation experiments, the results showed that the proposed algorithm improved convergence and distribution indicators by at least 84% and 76% respectively compared to NSGA-II, a classic multi-objective evolutionary algorithm, underscoring its superior search performance.