一种改进惯性权重的变异微粒群优化算法

An Improved Inertia Weight Mutation Particle Swarm Optimization

针对微粒群优化算法的早熟收敛和进化后期收敛速度慢等问题,提出了一种改进惯性权重的变异微粒群优化算法。在算法运行过程中,对适应度值不同的微粒赋予不同的惯性权重,使算法既具有良好的空间探索能力又有良好的局部挖掘能力;在群体最优信息陷入停滞时引入变异算子,对聚集在局部最优微粒附近的微粒的位置和速度进行变异操作,使算法摆脱局部最优点的束缚。对4种典型函数的测试结果表明,新算法的全局搜索能力和收敛速度都得到了提高,并且能够有效避免早熟收敛问题。

Proposes an improved inertia weight mutation particle swarm optimization to solve the premature convergence problem, and to avoid the slow - convergence in the later convergence phase. When running the algorithm, different inertia weight values are given to particles according to their fitness. Thus the algorithm is engaged with both good exploration ability and good exploitation ability. When the optimum information of the swarm is stagnant, mutation operator is introduced to change the location and speed of the particles which are close to the local optimum position, and thus to reduce the possibility of trapping at the local optimum. According to the experimental resuits using four typical functions, the global searching ability and the speed of convergence of the new algorithm are both improved, and the premature convergence problem is effectively avoided.