Application of linearised hierarchical models to indicated torque modelling for a turbocharged engine
by Mark Cary; Byron Mason; Peter Schaal
International Journal of Powertrains (IJPT), Vol. 5, No. 4, 2016

Abstract: This paper shows the application of linearised hierarchical models to the estimation of indicated torque data obtained from a large scale engine mapping experiment conducted on a turbocharged spark-ignition engine. Unlike previous studies that have utilised two-stage regression techniques for analysis, the use of linearised methods provides a framework to directly address the issues of sparseness of sweep-specific data and mixed effects modelling. In addition, spline models are presented at both levels of the model hierarchy that possess the required smoothness, are capable of capturing physical behaviour and simultaneously yield models sufficiently accurate for calibration work. The paper considers, at length, the required model fitting procedures which are founded on an iterative generalised least-squares approach. Further, a model building case study is presented addressing the issue of which factors should be modelled as fixed or mixed. This utilises information criteria to identify the most parsimonious model. Finally, the fit provided by the model to the data is demonstrated, which is seen to satisfy the engineering measure of success applicable to the application.

Online publication date: Thu, 26-Jan-2017

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Powertrains (IJPT):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com