基于改进模糊K-M法的岩溶塌陷危险性预测-《郑州大学学报(工学版)》

文章信息/Info

Title:: The Application of the Improved Fuzzy k-modes Algorithm to Forecastthe Karst Collapse Hazard

Author(s):: Feng Yong; Chen Pengju; 1. Civil engineering school of Henan Wniversity of Technology, Zhengzhou 450001 , China; 2. Geology Exploration institutionof Zhejiang Province,Hangzhou 310005,China

Keywords:: karst collapse; improved fuzzy k-modes algorithm; hazard forecast

摘要:: 岩溶塌陷受诸多因素影响,而这些影响因素在进行安全性评价中通常用定性的数据形式给出,尽管定量预测的数学理论已相当成熟,但是处理这种大量定性数据问题却有很大的局限性.因此,研究处理定性数据的预测方法对于开展岩溶塌陷的危险性具有重要意义.模糊K-M算法采用各类中各属性值的频率作为聚类的中心,通过各属性的权重来修正目标间的距离,采用非随机的方法选取初始聚类的中心,有效地保证了复杂定性数据计算结果的精确性和可靠性.在分析武汉市岩溶塌陷影响因素的基础上,利用此方法,对武汉市岩溶塌陷危险性进行研究,结果表明该种算法具有计算过程方便及收敛速度快等优点,值得在岩溶塌陷危险性预测中推广应用.

Abstract:: The influence factors of karst collapse are very complex, and they are usually be expressed by cate.gorical data. Although the mathematical theory of quantitative forecasting is quite mature, it has many limitations in processing large scale qualitative data. Therefore, it is necessary to introduce and apply processingqualitative data to forecast the ground collapse risk in karst region, For the improved fuzzy k-modes algorithm,the frequency of each attribute value of each attribute is taken as the cluster’s center, The distance of objectscan be modified by the weight of each attribute. A non-random method is applied to choose the clustering cen-ters , which can ensure the clustering results of complex qualitative data accuracy and reliability. Therefore, inthis paper, based on analysis on the influenee factors in Wuhan karst collapses, the improved fuzzy k-modesalgorithm was used to forecast the hazard potentiality of karst collapse in Wuhan city and good results havebeen obtained, It was found that the application of this model could get good effictiveness and this modelshould be adopted widely.