一种动态惯性权重的自适应粒子群优化算法

在标准粒子群算法中,权重过大导致最优点的搜寻能力降低,不能适应复杂的非线性优化搜索过程,动态惯性权重的自适应粒子群算法(APSO)解决了这一问题。在该算法中,粒子群中所有粒子适应度的整体变化可以跟踪粒子群的状态,在每次迭代时,算法可根据粒子的适应度变化动态改变惯性权重,从而使算法具有动态自适应性。通过对几种典型函数的测试结果表明,APSO算法的收敛速度和收敛精度明显优于LDW算法,从而提高了算法的性能。

Adaptive Particle Swarm Optimization Algorithm with Dynamically Changing Inertia Weight

A new adaptive particle swarm Optimization algorithm with dynamically changing inertia weight(APSO) is presented to solve the problem that the linearly decreasing weight(LDW) of the particle swarm algorithm cannot find out the optimum values sufficiently with high inertia weight.The algorithm also cannot adapt to the complex and nonlinear optimization process.During the running time,the inertia weight is determined by the particle's fitness that makes the algorithm become dynamical and adaptive.The experimental results show that the new algorithm not only has great advantage of convergence speed and accuracy,but also can increase the convergence accuracy.