A Parallel Computing Optimization Algorithm Based on Many-Objective

It is difficult for many-objective optimization problem to solve the problem of excessive consumption time effectively and maintain the balance of convergence and distribution of the solution. Aiming at this problem, a many objective parallel optimization based on external archive is proposed. The main idea is to design a many-objective selection mechanism penalty-based boundary intersection to enhance the selection pressure of the algorithm. Then an external archive is established to store the solution of convergence and diversity and guide the population towards evolve to the real pareto front. At the same time, in order to reinforce the diversity of the population, evolutionary search is further run on external archive. In addition, the idea of parallel is established to optimize the time. To test the performance of the algorithm, the algorithm is applied to some benchmark problems with 4-15 objectives and compared against other state-of-the-art algorithms. The simulation results demonstrate that the algorithm can effectively guarantee the convergence and diversity of solutions and shorten the time.