Materials Algorithms Project
Data Library
Data Library MAP_DATA_WELDHOT
- Provenance of code.
- Purpose of code.
- Description of subroutine's operation.
- References.
- Any additional information.
- Keywords.
- Download source code.
- Links.
Provenance of Source Code
Kazutoshi Ichikawa,
Nippon Steel Corporation,
Welding and Joining Research Center,
20-1 Shintomi, Futsu, Chiba 293-8511,
Japan
Top |
Next
Purpose
This library contains a number of files necessary to use a model for the estimation of solidification cracking in low-alloy steel welds using neural
network analysis, as described in [1].
Top |
Next |
Prev
Description
The library contains eight distinct data files :-
- wk2 contains the weights that are needed to
reproduce the results in the best model with the smallest marginalised
test error [1].
- wk2.lu contains the data that are needed to
reproduce the error bars in the best model [1].
- wk5 contains the weights that are needed to
reproduce the results in the second best model
[1].
- wk5.lu contains the data that are needed to
reproduce the error bars in the second
best model [1].
- wk3 contains the weights that are needed to
reproduce the results in the third best model
[1].
- wk3.lu contains the data that are needed to
reproduce the error bars in third best model
[1].
- wk4 contains the weights that are needed to
reproduce the results in the fourth best model
[1].
- wk4.lu contains the data that are needed to
reproduce the error bars in the fourth best model
[1].
The details of the neural network method are in [2]. The data
are arranged precisely as described in [1].
Top |
Next |
Prev
References
-
K. Ichikawa, H. K. D. H. Bhadeshia and D.J.C. MacKay, Model for Solidification Cracking in Low-Alloy Steel Weld Metals,Science and Technology of Welding and Joining, 1, (1996) 43-50.
-
D.J.C. MacKay, Bayesian non-linear modelling with neural networks, Mathematical Modelling of Weld Phenomena III, editors H. Cerjak and H.K.D.H. Bhadeshia, Institute of Materials, (1997) 359-389.
Top |
Next |
Prev
Further Comments
None.
Top |
Next |
Prev
materials, data, neural, network, low-alloy, steel, weld, weld deposit, crack, cracking, solidification
Top |
Next |
Prev
Download data
Download data library
Top |
Prev