论文部分内容阅读
提出一种搜索空间自适应的自适应粒子群优化算法.该算法对不同等级的粒子适应值采取不同的惯性权重,并随着算法的迭代不断缩小粒子群的搜索空间.同时,选择当前代的较优部分粒子直接进入下一代,其他粒子通过在缩小的搜索空间内随机生成,加快了种群收敛速度,同时又能使种群不断跳出局部最优解.几种典型函数的仿真实验表明,该算法在收敛速度和收敛精度上均较标准粒子群优化算法和普通自适应粒子群优化算法有明显提高.
A adaptive particle swarm optimization algorithm with search space adaptive is proposed. The algorithm takes different inertia weights for different levels of particle fitness and reduces the searching space of particle swarm as the algorithm iterates. At the same time, The optimal part of the particles directly into the next generation, other particles by randomly generated in the reduced search space to speed up the population convergence rate, while allowing the population continue to jump out of the local optimal solution.Some typical functions of the simulation experiments show that the algorithm Compared with the standard particle swarm optimization algorithm and the ordinary adaptive particle swarm optimization algorithm, the convergence speed and the convergence precision are obviously improved.