基于主成分分析-Logistic模型的边坡稳定性分析

摘要

提出一种基于PCA-Logistic模型算法,并对边坡的稳定性进行预测分析.先用PCA对样本数据进行处理,有效控制稳定性因素的计算数量.选出边坡重度、内聚力、内摩擦角、边坡角4个主成成分进行Logistic建模分析.得出其Logistic回归模型,并运用于边坡稳定性实例分析,与用Logistic回归模型确定权重的模糊综合法进行对比.研究结果表明:PCA-Logistic模型能有效地处理样本数据,减少不必要的分析过程,计算得出的边坡稳定性状态与实际情况相符;并与Logistic回归模型确定权重的模糊综合法进行对比,验证其计算结果的准确性,高效性.该方法可在工程地质、采矿工程、经济等众多领域应用与推广.%A PCA-Logistic model algorithm is proposed , and the stability of the slope is predicted and analyzed . The PCA method is used to preprocess the sample data and control the calculation of the stability factors .The four principal components of rock weight , rock cohesion , internal friction angle and slope angle are selected for Logistic modeling analysis .The Logistic regression model is applied to slope stability analysis , and compared with the fuzzy comprehensive method of Logistic regression model .The results show that the PCA-Logistic model can effectively deal with the sample data and reduce the unnecessary analysis process .The calculated slope stability state is in ac-cordance with the actual situation , and it is compared with the fuzzy synthesis method of Logistic regression model . The accuracy and efficiency of the method can be applied and popularized in many fields such as engineering geolo -gy, mining engineering , economy and so on .