基于速度交流的共生多种群粒子群算法

为了提高粒子群算法的求解精度,改善算法的搜索性能,提出一种基于速度交流的共生多种群粒子群算法(SMPSO)。该算法采用速度交流机制划分整个从种群为多个子种群,负责解空间的全局搜索,将获得的最优信息分享给主种群;主种群综合从种群与自身最优经验,负责局部深度优化,获得最优信息反馈给从种群,从而建立主从群间的共生关系,实现解空间的充分搜索。迭代后期,在主种群中引入自适应变异策略,提高算法跳出局部最优的能力。将提出的SMPSO算法应用于基准测试函数中,与其它改进的PSO算法进行比较。实验结果表明,SMPSO算法在求解精度、搜索能力、稳定性等方面均有较大的提高。

Symbiosis Multi-population Particle Swarm Optimization Algorithm Based on Velocity Communication

In order to improve the accuracy and the search performance of the particle swarm optimization algorithm, symbiotic multi-population particle swarm optimization algorithm based on velocity communication(SMPSO) was proposed.The whole slave population was divided into multiple sub-populations by using the speed communication mechanism for the global search of the solution space, and the optimal state of the slave population was sent to the master population.The master population comprehensive the slave population experience and its own experience for the local deep optimization, and the best information was sent to the slave population, establishing symbiotic relationship between the master-slave population and achieving full search of the solution space.In the latter part of the iteration, the master population was combined with the adaptive mutation strategy, increasing the ability of the algorithm to jump out of the local optimum.The proposed SMPSO algorithm was applied to the benchmark function and compared with other improved PSO algorithms.The experimental results showed that the SMPSO algorithm has great improvement in solving accuracy and search ability.