粒子群优化算法的运动学分析-西南大学.docVIP

PSO优化算法的微粒寻优过程移动规律研究 邓怀勇1，吴开贵2 (1.重庆水利电力职业技术学院，机械与电子信息系，重庆永川 402160 ; 2. 重庆大学，计算机学院算法简化模型，了寻优过程的运动学方程给出了基本粒子群优化算法的寻优运动规律。，简化模型中粒子的寻优运动规律当前最佳适应值为平衡点振动基本中粒子的寻优运动规律粒子以随机的角频率和振幅围绕平衡点振动粒子群优化运动学article Movement Rules of PSO Optimization Algorithm Deng Huaiyong1，Wu Kaigui2 (1. ChongQing Water Resources and Electric Engineering College, Chongqing 402160; 2. Chongqing University，Chongqing 401135) In the optimization process of Particle Swarm Optimization (PSO) algorithm, for the problem of lacking of knowledge about particle movement rule, meanwhile, in order to better expand and apply the particle swarm optimization (PSO) algorithm, based on a simplified model of PSO algorithm, In process of the particle optimization, the kinematical equation about particle vibrations around the equilibrium point is established, Then the relationship between parameter selection and angular frequency of vibration and amplitude of the movement of particles is given, and the particle optimization movement rules of the basic PSO algorithm are provided. Simulation results show that, 1 the particle optimization movement rule of the simplified PSO model is that the particle is around the vibration with the current best fitness value as the equilibrium point. 2 The particle optimization movement rule of the basic PSO model is that particles vibrate around the equilibrium point with random angular frequencies and amplitudes. These two rules fill the gaps in the understanding of the moving rules of the particle optimization process, which is helpful to popularize and improve the PSO algorithm so as to make it suitable for solving the practical optimization problem. Keywords Particle Swarm Optimization Algorithm; particle optimization; kinematics; movement rule 0引言 Kennedy和Eberhart粒子群优化算法PSO( Particle Swarm Optimization)交叉、变异算子，。PSOPSO算法的各种研究已经广泛展开，不管是算法发展以及实际应用方面都在不断产生新的研究成果。 对于算法发展方面：文献3采用动态策略的惯性权重动态调整传统的线性递减的惯性权重方法一种自适应惯性权重的均值粒子群优化算法该算法在收敛速度及精度都优于传统方法变邻域搜索改进的冲粒子群算法该算法使得收敛速度加快文献模拟退火与粒子群优化组合改进粒子群优化