一种基于种群速度的自适应粒子群算法  

Adaptive particle swarm optimization algorithm based on population velocity

作　　者：张顶学[1] 廖锐全[1] 

机构地区：[1]长江大学石油工程学院,湖北荆州434023

出　　处：《控制与决策》2009年第8期1257-1260,1265,共5页Control and Decision

基　　金：国家自然科学基金项目(60873021);湖北省高等学校优秀中青年团队计划项目(T200803)

摘　　要：分析了粒子群算法的收敛性,指出早熟是由于粒子速度降低而失去继续搜索可行解的能力.进而提出一种基于种群速度动态改变惯性权重的粒子群算法,该算法以种群粒子平均速度为信息动态改变惯性权重,避免了粒子速度过早接近0.通过5个标准测试函数的仿真实验并与其他算法相比,结果表明该算法在进化中期能很好地保持种群多样性,有效地改善算法的平均最优值和成功率.The convergence of particle swarm optimization （PSO） algorithm is analyzed. Its premature convergence is due to the decrease of the velocity of particles in search space. An adaptive PSO algorithm with dynamical changing inertia weight based on population velocity is proposed. The information defined as the average absolute value of velocity of all particles is defined as information to change the inertia weight dynamicly, which can avoid the velocity closed to 0 in the early search part. The simulation results show that the algorithm has better probability of finding global optimum and mean best value and can maintain the population diversity in the process of evolution.

关 键 词：粒子群优化算法 惯性权重 种群多样性 收敛性 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]