This content is not included in
your SAE MOBILUS subscription, or you are not logged in.
Correlation of Axle Build Parameters to End-of-Line NVH Test Performance Part II: Multivariate Regression Analysis
Technical Paper
2012-01-0728
ISSN: 0148-7191, e-ISSN: 2688-3627
Annotation ability available
Sector:
Language:
English
Abstract
The second part of a detailed examination of multivariate correlation of several axle assembly and component parameters to the assembly NVH performance (vibration) measured at the end of the assembly process is presented focusing on the multivariate linear regression analysis. The study is based on test results and measurements acquired from multiple axle assemblies built with the same hypoid gearset, thus effectively eliminating the affect of gearset variation on the test result. Several major components within the axle are considered including the differential housing (that controls wheel differentiation during turns), the axle housing, and several assembly parameters. Details of the multivariate regression include formulation of the linear regression model, model refinements through analysis of subsets of the variables, tests of significance and residual analysis. The first part of this work presented the analysis preparing the data for the regression analysis including checks for normality, checks for outliers and transformations.
Recommended Content
Citation
Meinhardt, G. and sengupta, S., "Correlation of Axle Build Parameters to End-of-Line NVH Test Performance Part II: Multivariate Regression Analysis," SAE Technical Paper 2012-01-0728, 2012, https://doi.org/10.4271/2012-01-0728.Also In
References
- Breiman, L. Friedman, J. “Predicting Multivariate Responses in Multiple Linear Regression,” Journal of the Royal Statistical Society 1997 1 3 54
- Olejnik, S. et al “Using Wherry's Adjusted R2 and Mallow's Cp for Model Selection from all Possible Regressions,” Journal of Experimental Education 2000 68 4 365 380
- Johnson, R. Wichern, D. Applied Multivariate Statistical Analysis 6 th Prentice Hall 2007
- Mendenhall and Sincich Statistics for Engineering and the Sciences 4 th Prentice Hall 1995
- Oja, H. Multivariate Nonparametric Methods with R: An Approach based on Spatial Signs and Ranks , Chapter 13 - Multivariate Linear Regression Lecture Notes in Statistics 2010 199 183 200
- Sun, Z. Steyer, G. Meinhardt, G. Ranek, M. “NVH Robustness Design of Axle Systems,” SAE Technical Paper 2003-01-1492 2003 10.4271/2003-01-1492
- Sun, Z. Ranek, M. Voight, M. Steyer, G. “Variation Reduction of Axle System NVH,” SAE Technical Paper 2005-01-2309 2005 10.4271/2005-01-2309