带变异算子的非线性惯性权重PSO算法

为了克服粒子群优化算法容易陷入局部最优、早熟收敛的缺点，提出了一种带有变异算子的非线性惯性权重粒子群优化算法。该算法以粒子群算法为基础，首先采用非线性递减策略对惯性权重进行调整，平衡粒子群优化算法的全局和局部搜索能力。当出现早熟收敛时，再引入变异算子，对群体粒子的最优解做随机扰动提高算法跳出局部极值的能力。用三种经典测试函数进行测试，试验结果表明，改进算法与粒子群算法相比，能够摆脱局部最优，得到全局最优解，同时具有较高的收敛精度和较快的收敛速度。

A Nonlinear Inertia Weight Particle Swarm Optimization Algorithm with Mutation Operator

In order to overcome the shortcomings that standard particle swarm algorithm is easy to fall into local optima and premature convergence, a nonlinear inertia weight particle swarm optimization improved algorithm with mutation operator is proposed. On the basis of the PSO algorithm,firstly the new algorithm introduces nonlinear decreasing strategy to adjust the weight of inertia,balances the parti- cle swarm optimization global and local capabilities. When the optimization is in premature convergence, introduce mutation operator to do random perturbations for the optimal solution of the particle group to improve the ability of the algorithm to jump out of local extreme. Three benchmark functions are tested and the experimental results show that the improved algorithm is able to get rid of local extreme,get the global optimal solution, but also has higher convergence precision and convergence speed than the particle swarm algorithm.