【摘要】 为了进一步提高粒子群算法的性能,提出了一种新的群体智能优化算法——带高斯扰动和协同寻优的蝙蝠粒子群混合算法。该混合算法利用蝙蝠个体脉冲的回声定位对最优粒子gbest进行高斯扰动而产生一个局部解,把该局部解加到蝙蝠种群中,然后根据局部解的位置优劣与蝙蝠个体产生的响度来更新粒子群。在寻优过程中,对gbest进行高斯扰动增加了种群的多样性而避免粒子群过快陷入局部最优,并且加强了蝙蝠种群与粒子群的信息交互,协同寻优。与蝙蝠算法、标准粒子群算法、烟花算法、带高斯扰动的粒子群算法、粒子群差分算法相比,带高斯扰动和协同寻优的蝙蝠粒子群混合算法的总体性能优于其他5种算法。更多还原

【Abstract】 In order to further improve the performance of particle swarm optimization algorithm,a new swarm intelligent optimization algorithm named hybrid Bat and Particle Swarm Optimization algorithm with Gaussian disturbance and cooperation optimization is proposed. The new algorithm uses the echolocation of the individual bat particle’s pulse to produce a local solution to the optimal particle of gbestby Gaussian perturbation. The local solution is added to the bat population,and then the particle swarm is updated according to the location of the local solution and the loudness of the individual bat. In the process of optimization,Gaussian disturbance increases the diversity of the population and avoids the particle swarm into the local optimal too fast,and strengthens the information interaction and collaborative optimization between the bat population and the particle swarm. Compared with bat algorithm,standard particle swarm algorithm,fireworks algorithm,PSO with Gaussian mutation,and PSO-DE,the presented hybrid algorithm has better overall performance.更多还原