对微粒群算法结构的改进方案有很多种，对其可分类为：采用多个子种群；改进微粒学习对象的选取策略；修改微粒更新迭代公式；修改速度更新策略；修改速度限制方法、位置限制方法和动态确定搜索空间；与其他搜索技术相结合；以及针对多模问题所作的改进。-Structure of the particle swarm algorithm to improve the program there are many, its can be classified as: the use of multiple sub-populations　improve the selection of learning objects strategy particles　modified particles iterative update formula　modify the speed update strategy　modified method of speed limits, position limits methods, and dynamically determines the search space　and other search technologies　and multi-mode problem for the improvements made.