Statistical modelling of composition and processing parameters for alloy development

Abstract

We propose the use of regression models as a tool to reduce time and cost associated with the development and selection of new metallic alloys. A multiple regression model is developed which can accurately predict tensile yield strength of high strength low alloy steel based on its chemical composition and processing parameters. Quantile regression is used to model the fracture toughness response as measured by Charpy V-Notch (CVN) values, which exhibits substantial variability and is therefore not usefully modelled via standard regression with its focus on the mean. Using Monte-Carlo simulation, we determine that the three CVN values corresponding to each steel specimen can be plausibly modelled as observations from the 20th, 50th and 80th percentiles of the CVN distribution. Separate quantile regression models fitted at each of these percentile levels prove sufficiently accurate for ranking steels and selecting the best combinations of composition and processing parameters.

Modelling of Charpy V test rejection probability

“…Another promising approach for predicting CVT rejections is conditional distribution modelling with quantile regression. 6 The model will be implemented into a graphical simulation tool that is in daily use in the product planning department and already contains other mechanical property models. 17 The simulation tool is utilised to plan composition and production settings for product modifications and new products and to maintain the regulations for the production methods of existing products.…”

“…The test was performed on only one grade of steel and only at one test temperature. 6 Methods for joint modelling of mean and dispersion in different industrial applications have been studied widely, [7][8][9] and the typical method is to perform heteroscedastic regression with generalised linear models.…”

New Developments in Uncertainty: Linking Risk Management, Reliability, Statistics and Stochastic Optimization

“…There could be many applications of these ideas, and much remains to be explored and developed. Some related research in special cases, largely concerned with quantile regression, can be seen in Trindade and Uryasev [2006a], Trindade and Uryasev [2006b] and Golodnikov et al [2007]; see also Samson et al [2009] for further motivation.…”

‘Symbiotic’ data-driven modelling for the accurate prediction of mechanical properties of alloy steels

“…Over the years several models have been proposed in the open literature for the characterization of behaviour relating to thermomechanical processing ranging from physical models [8], [11], to data-driven model architectures [4]- [6], [13]- [15]. The one common denominator linking all these models related to their ability to deal with the accuracy of the predictions 978-1-4244-5164-7/10/$26.00 ©2010 IEEE which represents their "raison d'être".…”