A neural network technique trained within a Bayesian framework has been applied to the analysis of the yield strength, ultimate tensile strength and percent elongation of mechanicallly alloyed ODS ferritic steels. The database was complied with information from the published literature, consisting of variables known to be important in influencing mechanical properties. The analysis has produced patterns which are metallurgically reasonable and which permit the quatitative estimation of mechanical properties together with an indication of confidence limits.
Materials Science and Technology, 14 (1998) 793-809.
Download review of Neural Networks in Materials Science
The components of the yield strength of MA956, a mechanically alloyed oxide-dispersion strengthened iron base superalloy, have been investigated quantitatively. It is found that much of the difference between the recrystallised and unrecrystallised forms can be explained in terms of the grain structure. The contribution from dispersion strengthening has been estimated using dislocation theory and has been demonstrated to be consistent with that measured experimentally. The temperature dependence of the yield strength has also been studied; some of the effects observed in the range 500-600 Centigrade can be attributed to the change in the intrinsic strength of pure, annealed iron.
Materials Science and Technology, 14 (1998) 1221-1226.
|
|
|
| Superalloys | Titanium | Bainite | Martensite | WidmanstŠtten ferrite |
| Cast iron | Welding | Allotriomorphic ferrite | Movies | Slides |
| Neural Networks | Creep | Mechanicallly Alloyed | Theses | Retained Austenite |
| PT Group Home | Materials Algorithms |