论文部分内容阅读
影响黄土湿陷性等级的因子之间存在着相关性,针对这一问题,利用聚类分析和因子分析理论研究了因子之间的相关性,并建立了分析模型。选用天然重度、干重度、孔隙比、含水率、塑性指数5项指标作为评价因子,以湿陷系数作为评价指标,通过聚类分析和因子分析表明黄土的天然重度、干重度、孔隙比对黄土湿陷性等级影响具有共性,可归为一类因子;含水率和塑性指数分别为一类,这样就将相关的原始因子转换为相互独立的因子。将聚类分析和因子分析引入黄土湿陷性系数等级评价中,不仅保留了原始因子的信息,而且还消除了多重共线性的影响,减小了误差,提高了分析结果的准确性。工程实例分析表明,提出的分析模型与实际工程相吻合,具有重要的工程实际意义。
In order to solve this problem, the correlation between factors is studied by cluster analysis and factor analysis theory, and the analysis model is established. Five indicators of natural heavy degree, dry weight, porosity, water cut, plasticity index were selected as evaluation factors and collapsibility coefficient as evaluation index. Cluster analysis and factor analysis showed that natural loess, dry weight and void ratio of loess Collapsible grade effects are common and can be classified as a category of factors; water content and plasticity index of a class, so that the relevant original factor is converted to independent factors. Introducing cluster analysis and factor analysis into the evaluation of collapsibility coefficient of loess, not only retain the information of the original factor, but also eliminate the influence of multicollinearity, reduce the error and improve the accuracy of the analysis results. The analysis of engineering examples shows that the proposed analysis model is in line with the actual project, which has important practical significance.