Quantum dynamic mechanism-based multi-objective evolutionary algorithm and performance analysis

Abstract

A novel Self-organizing Quantum Evolutionary Algorithm for Multi-objective optimization(MSQEA) is proposed. The technique for improving the performance of MSQEA has been described. By using self-organizing co-evolution strategy each subpopulation can obtain more optimal solutions. Because of the quantum dynamic mechanism all the subpopulations may move concurrently in a force-field until all of them reach their equilibrium states. Self-organizing algorithm has advantages in terms of the adaptability; reliability and the learning ability over traditional organizing algorithm, so the solution quality is improved. 0/1 Multi-objective knapsack problem simulation results demonstrate the superiority of MSQEA in this paper.