Prediction of martensite start temperature
Source abstract
Abstract
Methods of predicting the martensite start temperature as a function of composition have been evaluated. A technique has been demonstrated which improves the performance of linear regression models by applying the concept of a committee borrowed from more sophisticated empirical techniques. These linear regression models, neural network models, thermodynamic models and a hybrid thermodynamic–neural network model are tested using various assessment parameters. The thermodynamic model has the best performance when tested within a typical range of the input space. Bayesian neural networks possess the advantage that their predictions are naturally accompanied by an estimate of uncertainty and they can have the best performance when this is considered. Incorporating the thermodynamic model into a neural network combined the advantages of the two methods.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Microstructure and Mechanical Properties of Steels · High Temperature Alloys and Creep · Metallurgy and Material Forming
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Materials Science and Metallurgy, University of Cambridge, 27 Charles Babbage Road, Cambridge CB3 0FS, UK
References (34)
Show source-provided references
1. Payson P · Trans. ASM · 1944 · 33 · 261
2. Carapella LA · Met. Prog. · 1944 · 46 · 108
3. Rowland ES · Trans. ASM · 1946 · 37 · 27
4. Grange RA · Trans. AIME · 1946 · 167 · 467
5. Nehrenberg AE · Trans. AIME · 1956 · 167 · 494
6. Steven W · J. Iron Steel Inst. · 1956 · 183 · 349
7. Andrews KW · J. Iron Steel Inst. · 1965 · 203 · 721
8. Kunitake T · Sumitomo Search · 1969 · 2 · 18
9. Tamura I: ‘Steel material study on the strength’; 1970, Tokyo, Nikkan Kogyo Shinbun Ltd. In Japanese.
10. Monma K: ‘Tekko-zairyo-gaku’, chap. Martensite, 158–163; 1972, Jikkyo-Syuppan Corp, Tokyo. In Japanese.
11. Eldis GT: ‘A critical review of data sources for isothermal transformation and continuous cooling transformation diagrams’, Proc. Symp. on ‘The hardenability concepts with applications to steel’, 126–157; 1977, The Metallurgical Society of AIME, Warrendale, PA.
12. 10.1007/BF0264332710.1007/BF02643327 ↗
13. Pickering FB: ‘Physical metallurgical development of stainless steels’, in ‘Stainless steels ’84’, 2–28; 1985, London, The Institute of Metals.
14. 10.1002/srin.19960549810.1002/srin.199605498 ↗
15. Kunitake T · J. Jpn Soc. Heat Treat. · 2001 · 41 · 164
16. 10.2320/jinstmet1952.34.3_29110.2320/jinstmet1952.34.3_291 ↗
17. 10.2355/isijinternational.39.96610.2355/isijinternational.39.966 ↗
18. Vermeulen W · Ironmak. Steelmak. · 1996 · 23 · 433
19. 10.2320/matertrans1989.41.76110.2320/matertrans1989.41.761 ↗
20. 10.1016/j.commatsci.2005.01.00210.1016/j.commatsci.2005.01.002 ↗
21. 10.1016/0956-7151(94)90468-510.1016/0956-7151(94)90468-5 ↗
22. 10.1016/0956-7151(94)90469-310.1016/0956-7151(94)90469-3 ↗
23. 10.1016/j.commatsci.2005.01.00110.1016/j.commatsci.2005.01.001 ↗
24. Sourmail T: ‘MAP_DATA_STEEL_MS_2004’, available at: http://www.msm.cam.ac.uk/map/ (accessed 11 November 2014).
25. MacKay DJC: ‘Bayesian methods for adaptive models’, PhD thesis, Caltech, Pasadena, CA, USA, 1992.
26. MacKay DJC: ‘Bayesian methods for neural networks: Theory and applications’, Cavendish Laboratory, 1995.
27. MacKay DJC · 1997 · 359
28. 10.1016/S0921-5093(01)00913-310.1016/S0921-5093(01)00913-3 ↗
29. 10.1002/sam.1001810.1002/sam.10018 ↗
30. 10.2355/isijinternational.42.89410.2355/isijinternational.42.894 ↗
31. 10.1007/978-1-4612-0745-010.1007/978-1-4612-0745-0 ↗
32. 10.1179/02670830222500206510.1179/026708302225002065 ↗
33. 10.1179/174328407X17685710.1179/174328407X176857 ↗
34. Peet M: MAP_MS_MODELS_2014, Materials Algorithms Project, available at: http://www.msm.cam.ac.uk/map/ (accessed 12 December 2014).