多元线性回归法在车用汽油品质影响因素分析中的应用  

Application of multiple linear regression method in the analysis of influencing factors of motor gasoline quality

作　　者：徐晓敏[1] 严雨琪 XU Xiaomin;YAN Yuqi(School of Business,Shanghai Dianji University,Shanghai 201306,China)

机构地区：[1]上海电机学院商学院,上海201306

出　　处：《上海电机学院学报》2022年第2期112-117,共6页Journal of Shanghai Dianji University

基　　金：上海电机学院学科建设资助项目(16YSXK02)。

摘　　要：为改善车用汽油品质,首先,用因果图对影响车用汽油品质的潜在因素进行分析,根据要因确认表,得出影响车用汽油品质的关键因素——辛烷值;其次,分析辛烷值的各个参数,由此可知,影响辛烷值的关键参数为苯、芳烃、烯烃和氧的含量;然后,利用多元线性回归中的逐步回归法消除自变量间的多重共线性,从而得到回归方程,确定影响车用汽油品质的关键参数;最后,为改善车用汽油品质提出建议并制定方案。To improve the quality of motor gasoline, first, the potential factors that affect the quality of motor gasoline are analyzed by the fishbone diagram. The key factor affecting the quality of motor gasoline, i.e. octane number, is obtained according to the key factor confirmation table. Second, it can be seen that the key parameters affecting octane number are benzene content, aromatics content, olefin content and oxygen content by analyzing the parameters of octane number. Then, the stepwise regression method in multiple linear regression is used to eliminate the multicollinearity between independent variables. The regression equation is obtained, and the key parameters affecting the quality of motor gasoline is determined.Finally, some suggestions and measures are put forward to improve the quality of motor gasoline.

关 键 词：车用汽油品质 因果图 多元线性回归 

分 类 号：U473.1[机械工程—车辆工程]