论文部分内容阅读
为解决地质采矿因素之间的相关性及信息重叠影响概率积分法参数的计算精度问题.采用主成分回归分析的方法,以43个典型观测站的实测数据作为样本,研究了概率积分法参数的计算方法.研究结果表明:通过主成分分析方法提取前5个主成分的累积方差贡献率已经达到91%,可用其代替原始的8个因素;该方法可以有效减少变量的个数,消除变量之间的相关性及信息重叠.采用提取的主成分进行多元线性回归计算参数的最大相对误差为8.9%,最大平均绝对误差百分率为5.4%;内部拟合误差及外部预测误差均较小,表明该方法的计算结果是准确可靠的.
In order to solve the problem of the correlation between geologic and mining factors and the influence of overlapping information on the accuracy of the parameters of the probability integral method, the principal component regression analysis method was used to take the measured data from 43 typical observatories as samples. The results show that the cumulative variance contribution rate of the first five principal components extracted by the principal component analysis method has reached 91%, which can be used instead of the original eight factors. This method can effectively reduce the number of variables and eliminate the variable The correlation and the overlap of information.The maximum relative error of the parameters calculated by multivariate linear regression using the extracted principal components is 8.9% and the maximum average absolute error percentage is 5.4%, both the internal fitting error and the external prediction error are small, The calculation results of the method are accurate and reliable.