基于鸟类群体响应的万有引力搜索算法及其在数据聚类中的应用

发布时间：2018-06-16 02:50

本文选题：万有引力搜索算法 + 启发式搜索算法　；参考：《山西科技》2016年06期

【摘要】：万有引力搜索算法是一个基于重力和质量动力学规律的随机搜索算法,广泛用于解决优化问题。为了增强原算法的性能,提出基于鸟类群体响应的万有引力搜索算法。该算法引入一个新的多样性勘探机制以增强它的多样性勘探能力,该机制通过3个主要步骤实现:初始化、最近邻域个体识别以及方向的改变。借助群体响应机制,提出的算法可以探索范围更广的搜索空间,并因此避免陷入局部次优解决方案。采用基准优化函数验证算法,并且应用该算法进行数据聚类。结果表明,提出的算法提高了原始GSA的多样性勘探性能,在优化函数实验和数据聚类中得到了更好的结果。
[Abstract]:The universal gravity search algorithm is a random search algorithm based on the law of gravity and mass dynamics, which is widely used to solve optimization problems. In order to improve the performance of the original algorithm, a universal gravity search algorithm based on bird population response was proposed. The algorithm introduces a new diversity exploration mechanism to enhance its diversity exploration ability. The mechanism is implemented through three main steps: initialization, recent neighborhood identification and direction change. With the help of group response mechanism, the proposed algorithm can explore a wider range of search space and thus avoid falling into local sub-optimal solution. The benchmark optimization function is used to verify the algorithm, and the algorithm is applied to data clustering. The results show that the proposed algorithm can improve the diversity exploration performance of the original GSA and obtain better results in optimization function experiments and data clustering.
【作者单位】：国网辽宁省电力有限公司信息通信分公司;太原理工大学新型传感器与智能控制教育部山西省重点实验室;
【基金】：国家自然科学青年基金(51405327) 山西省自然科学基金(2014011021-1,2014021024-1)
【分类号】：TP18

【相似文献】