Predicting the creep strength of austenitic stainless steels
Two things on this page:
- The most up to date database for the creep strength and creep rupture life of austenitic stainless steels, as a function of composition.
- The most up to date models to predict the creep rupture strength or creep rupture life of austenitic stainless steels.
The database:
- Open Office format.
- Excel format.
It contains a set of data (3100) which covers all the main AISI 300 series (type 304, 310, 316, 321, 347), their equivalent japanese SUS series, and a vast number of modified compositions found in the published literature.
The latest neural network model (July 2004)
The PC and Unix/Linux programs predict ln(-ln((y+2)/10)) as an output where y is the creep rupture life. This is automatic with Predictor.
- Standalone program for DOS/Windows: unzip, test by running model.exe, to predict on your compositions, edit test.dat where the columns are as in labels.txt.
- Source and script to install on linux/unix: extract (tar -zxf Ms_steels.tar.gz), run 'sh install', then run 'sh model.gen'. To predict on your compositions, edit test.dat where the columns are as in labels.txt.
- Package for Neuromat Predictor, simply select 'install package' in Neuromat Predictor.
Online calculations of creep rupture life and creep rupture stress can be done here.
Note: the model have been trained using confidential data which are not included in the distributed database.
Relevant publications:
-
Review: Precipitation in Creep-Resistant Austenitic Stainless
Steels, T. Sourmail, Mat.Sci.Tech. 2001: 17-1 p1-14
This was a commended entry of the 1999 IoM Litterature Survey competition. - A Neural Network Model of the Creep Strength of Austenitic Stainless Steels, T. Sourmail, H.K.D.H. Bhadeshia and D.J.C. MacKay Mat. Sci. Tech. 2002:18-6 p655-663
-
Microstructural evolution of two variants of NF709 during ageing at 1023 and
1073 K, T. Sourmail and H. K. D. H. Bhadeshia
Metall. Mater. Trans. A, 2005:36A, p23-34
- Other papers