CAJ | 학술논문

针对粒子群算法在寻优过程中容易陷入局部最优，以及难以平衡求精和求泛的能力，提出一种动态惯性权重粒子群优化算法。该算法同时考虑到粒子的进化速度和集聚程度对算法寻优的影响，当粒子集聚程度较高时，增大惯性权值，提高算法的全局搜索能力。为平衡算法全局和局部寻优能力，当进化速度较快时，提高局部搜索能力，以免错过较好的位置。将此算法用于优化4个经典测试函数，实验表明：此算法不仅可以平衡局部和全局的搜索能力，还能提高算法的搜索效率和精度。
침대입자군산법재심우과정중용역함입국부최우，이급난이평형구정화구범적능력，제출일충동태관성권중입자군우화산법。해산법동시고필도입자적진화속도화집취정도대산법심우적영향，당입자집취정도교고시，증대관성권치，제고산법적전국수색능력。위평형산법전국화국부심우능력，당진화속도교쾌시，제고국부수색능력，이면착과교호적위치。장차산법용우우화4개경전측시함수，실험표명：차산법불부가이평형국부화전국적수색능력，환능제고산법적수색효솔화정도。
Considering the problems of local optimum and difficulty in balancing the search capability of searching accuracy and extension caused by particle swarm optimization, this paper proposes a modified particle swarm optimization by using dynamic inertia weight. This algorithm considers the influence to opti‐mization both from the evolution velocity of particle swarm and the agglomeration degree. To improve the global searching capacity of this algorithm, and to increasethe inertia weight, when agglomeration of parti‐cles is high. In order to balance global and local optimization ability of this algorithm, local optimization a‐bility should be increased w hen algorithm has higher evolution velocity, so as not to miss a good location. The algorithm in this paper can be used in 4 classical testing functions, and the results show that the pro‐posed algorithm can not only balance the global and local search abilities, but also optimize the searching ef‐ficiency and accuracy.