论文部分内容阅读
应用主成分回归分析方法对新疆希尼尔水库大坝某断面的渗透压力预测进行分析研究。利用SPSS13.0软件进行多重共线性诊断后进行主成分分析,确定主成分的个数,将原自变量的主成分代替原自变量进行回归分析,建立主成分回归模型,并对该大坝渗透压力进行预报。对影响大坝渗透压力的各因子进行有效分离,既保留了原指标的绝大部分信息,又有主成分之间不相关的特点,弥补了最小二乘法回归无法有效识别和消除因子间多重相关性影响的不足。
The Principal Component Regression Analysis is used to analyze the prediction of infiltration pressure of a section of the Xinyilu Reservoir in Xinjiang. SPSS13.0 software was used to diagnose multicollinearity, then principal component analysis was carried out, the number of principal components was determined, the principal components of the original variables were used to perform regression analysis instead of the original variables, the principal component regression model was established, Pressure forecast. Effective separation of various factors affecting the seepage pressure of the dam not only retains most of the information of the original indicator but also does not correlate with the principal components to make up for the regression of the least square method can not effectively identify and eliminate multiple correlations between factors The lack of sexual influence.