基于QPSO和拥挤距离排序的多目标量子粒子群优化算法

Multi-objective quantum-behaved particle swarm optimization algorithm based on QPSO and crowding distance sorting

为了提高多目标优化算法的收敛性、分布性和减少算法的计算代价,提出一种基于量子行为特性的粒子群优化(QPSO)和拥挤距离排序的多目标量子粒子群优化(MOQPSO-CD)算法.MOQPSO-CD利用QPSO快速接近真实的Pareto最优解,同时引入高斯变异算子以增强解的多样性.采用拥挤距离排序的方法对外部存储器中最优解进行更新和维护,使得从中选择的具有全局最优的领导粒子能够引导粒子群最终找到真实的Pareto最优解.仿真结果表明,MOQPSO-CD具有更好的收敛性和更均匀的分布性.

For improving the convergence and distribution together with less computation cost of multi-objective optimization algorithm,a multi-objective quantum-behaved particle swarm optimization based on QPSO and crowding distance sorting（MOQPSO-CD） algorithm is proposed.MOQPSO-CD makes full use of QPSO to approximate the true Pareto optimal solutions quickly,and Gaussian mutation operator is introduced to enhance the diversity of solution.MOQPSO-CD updates and maintains the archived optimal solutions based on crowding distance sorting technique,whose purpose is making the leader particles with global optimal ability guide the particle swarm finding the true Pareto optimal solutions finally.Simulation results show that MOQPSO-CD has better convergence and distribution.