自适应变异的混合粒子群优化策略及其应用

提出了一种新的基于群体自适应变异和个体退火操作的混合粒子群优化（HPSO）算法．该算法将模拟退火过程引入到粒子群优化（PSO）之中，以PSO作为主体，先随机产生初始群体，并开始随机搜索产生新的个体．同时，使用自适应变异操作进行个体变异，对进化过的个体进行退火操作，以调整和优化群体．与模拟退火算法和基本PSO算法相比，HPSO保持了基本PSO算法简单、容易实现的特点，又能进行自适应变异．复杂函数优化和旅行商组合优化问题的实例验证表明，所提算法的全局收敛性较好，提高了摆脱局部最优的能力，有效避免了基本PSO算法的早熟问题．

Hybrid Particle Swarm Optimization Strategy with Adaptive Mutation and Its Applications

A novel hybrid particle swarm optimization (PSO) based on adaptive population mutation and individual annealing operation was developed.The simulated annealing(SA) operation was introduced into the PSO.Regarding PSO as the principal part of the hybrid strategy,initial colony was randomly generated firstly,and then new individuals were searched.Meanwhile,the adaptive mutation and annealing operation were used to adjust and optimize the population.Compared with SA and basic PSO,the hybrid PSO(HPSO) keeps the simple and convenient character of the standard PSO,and also carries on mutation adaptively.Experimental results for some complex function optimization and several TSP combination optimization problems show that the HPSO improves the global convergence ability and efficiently prevents the algorithm from the local optimization and early mature.