CAJ | 학술논문

针对标准微粒群易于收敛于局部最优解的问题，提出了一种通过添加粒子群搜索历史信息的改进微粒群算法。标准粒子群算法没有用到粒子群以往的迭代信息，再加上粒子群算法的快速收敛性，造成了算法易于收敛到局部最优解。改进的算法在运行时可以增加算法搜索空间，使算法更加平稳的收敛于全局最优解。通过用典型的Benchmarks函数进行模拟试验，实验结果证实了所提出的算法更加平稳，收敛速度更快。
침대표준미립군역우수렴우국부최우해적문제，제출료일충통과첨가입자군수색역사신식적개진미립군산법。표준입자군산법몰유용도입자군이왕적질대신식，재가상입자군산법적쾌속수렴성，조성료산법역우수렴도국부최우해。개진적산법재운행시가이증가산법수색공간，사산법경가평은적수렴우전국최우해。통과용전형적Benchmarks함수진행모의시험，실험결과증실료소제출적산법경가평은，수렴속도경쾌。
The standard particle swarm optimization algorithm is easily converging to the local optimal solution. The algorithm above seldom contains the historical information of particle swarm search path, adding the fast convergence of this algorithm, they make this optimization converging to the local optimal solution too fast. So an improved particle swarm optimization by adding the historical information of particle swarm search is presented, which adds the algorithm search space to make the algorithm converg?ing to global optimal solution steadily during the running time. Several classic Benchmarks functions are tested and the results show that the rate of convergence is faster and the algorithm is more stable.