边坡稳定性分类评价的同伦模糊C-均值聚类算法

针对单一模糊C-均值聚类算法对初始聚类中心初值敏感性问题，引入同伦理论，提出了同伦模糊C-均值聚类算法。以三峡库岸研究程度较高的36个边坡为对象，采用同伦模糊C-均值聚类算法对边坡的稳定性进行分类，研究边坡最佳分类级数和算法的收敛性、可靠性。边坡聚类结果研究表明，同伦模糊C-均值聚类算法对初始聚类中心的选取没有明显的依赖性，是一个具有全局最优解的聚类方法，其结果明显好于单一模糊C-均值聚类算法。

Study of classified evaluation of slope stability based on homotopy fuzzy C-mean clustering algorithm

In order to overcome the shortcoming of sensitivity to the original clustering center in ordinary fuzzy C-mean clustering algorithm,the homotopy theory is introduced to combine with this algorithm to establish a new method of homotopy fuzzy C-mean Clustering algorithm for evaluation of slope stability.Then the method is applied to studying stability of 36 slopes at banks of the Three Gorges Reservoir;and then some characteristics such as the best sample classification and convergence effect of homotopy fuzzy C-mean clustering algorithm are discussed.Clustering results show that the proposed method is a reasonable classification algorithm with global optimization and superior to the ordinary fuzzy C-mean clustering algorithm.