论文部分内容阅读
基于原始数据的成分数据变换和中心化对数比变换,对传统的主成分分析法进行“非线性”改进。运用改进的主成分分析法分析有调中时、无调中时、无调比对中时的影响,发现有调中时和无调比是影响中时的重要因素,改进的主成分分析法比传统方法具有原始变量降维效果好、主成分代表意义易解释的优点;选取办理车数、有调中转车数、货物列车到发列数、无调比、折角车流量和运用车保有量等6个因素指标,使用正、负、零关联度对影响因素与中时进行点关联分析,结果表明,办理车数、有调中转车数和货物列车到发列数等技术作业量的增长是中时减少的主要因素,提高无调比、减少折角车流和降低运用车保有量对压缩中时有重要作用;点关联分析比绝对值关联分析更能准确地分析各种因素对中时的影响规律。
Based on the transformation of composition data and the logarithm transformation of the original data, the traditional principal component analysis method is improved “nonlinearly”. The use of improved principal component analysis to analyze the effects of mid-tonal and non-tonal mid-tonic, we found that tonal-tonic and non-tonal ratio are the important factors influencing mid-tonal, and the improved principal component analysis Compared with the traditional method, the original variable has the advantage of reducing the dimensionality and the meaning of the principal component is easy to explain. The number of vehicles with transfer, the number of transferred transfers, the number of goods train to the serial number, And other six indicators, the use of positive, negative, zero correlation between the influencing factors and the time point-by-point analysis, the results show that for the number of cars, the number of transfer transfers and cargo trains to the number of train and other technical operations growth Is the main factor in the mid-time reduction, improve the non-transfer ratio, reduce the flow of cars and reduce the use of car ownership of compression plays an important role in time; point correlation analysis than absolute correlation analysis more accurately analyze the various factors on the time Affect the law.