Materials Algorithms Project
Program Library

Subroutine MAP_STEEL_WELD_CHARPY

1. Provenance of code.
2. Purpose of code.
3. Specification.
4. Description of subroutine's operation.
5. References.
6. Parameter descriptions.
7. Error indicators.
8. Accuracy estimate.
9. Any additional information.
10. Example of code
11. Auxiliary subroutines required.
12. Keywords.
13. Download source code.
14. Links.

Provenance of Source Code

Junhak Pak, May 2008, junhark@postech.ac.kr
Computational Metallurgy Laboratory,
Graduate Institute of Ferrous Technology,
POSTECH,
Pohang, Korea.

The neural network program was produced by:

David MacKay,
Cavendish Laboratory,
University of Cambridge,
Madingley Road,
Cambridge, CB3 0HE, U.K.

Top | Next

Purpose

To estimate Charpy toughness of steel weldmetal (manual metal arc, submerged arc or tungsten inert gas), as a function of chemical composition, heat input, interpass temperature, post weld heat treatment temperature and time.

Top | Next | Prev

Specification

Description

A program which can be used for predicting Charpy toughness of weldmetal (manual metal arc, submerged arc or tungsten inert gas). The input variables required to run the program are chemical composition, heat input, interpass temperature, post weld heat treatment temperature and time and test temperature for Charpy test. The model is in fact a committee of eight different models.

The downloadable package contains the following files and subdirectories (details may differ between LINUX and PC versions):

MINMAX

A text file containing the minimum and maximum limits of each input and output variable. This file is used to normalise and unnormalise the input and output data.

test.dat

An input text file containing the input variables used for predictions.

model.gen or model.exe

This is a unix shell file containing the command steps required to run the module. It can be executed by typing csh model.gen  at the command prompt. This shell file compiles and runs all the programs necessary for normalising the input data, executing the network for each model, unnormalising the output data and combining the results of each model to produce the final committee result.

spec.t1

A dynamic file, created by spec.ex, which contains information about the module and the number of data items being supplied. It is read by the program generate44.

norm_test.in

This is a text file which contains the normalised input variables. It is generated by the program normtest.for in subdirectory s.

generate44

This is the executable file for the neural network program. It reads the normalised input data file, norm_test.in, and uses the weight files in subdirectory c. The results are written to the temporary output file _out.

_ot, _out, _res, _sen

These files are created by generate44 and can be deleted.

Result

Contains the final un-normalised committee results for the predicted hardness.

SUBDIRECTORY s

spec.c

The source code for program spec.ex.

normtest.for

Program to normalise the data in test.dat and produce the normalised input file norm_test.in. It makes use of information read in from no_of_rows.dat and committee.dat.

gencom.for

This program uses the information in committee.dat and combines the predictions from the individual models, in subdirectory outprdt, to obtain an averaged value (committee prediction). The output (in normalised form) is written to com.dat.

treatout.for

Program to un-normalise the committee results in com.dat and write the output predictions to unnorm_com. This file is then renamed Result.

committee.dat

A text file containing the number of models to be used to form the committee result and the number of input variables. It is read by gencom.for, normtest.for and treatout.for.

SUBDIRECTORY c

_w_*f

The weights files for the different models.

_*.lu

Files containing information for calculating the size of the error bars for the different models.

_c_*

Files containing information about the perceived significance value [1] for each model.

_R_*

Files containing values for the noise, test error and log predictive error [1] for each model.

SUBDIRECTORY d

outran.x

A normalised output file which was created when developing the model. It is accessed by generate44 via spec.t1.

SUBDIRECTORY outprdt

out1, out2 etc.

The normalised output files for each model.

com.dat

The normalised output file containing the committee results. It is generated by gencom.for.

Top | Next | Prev

References

1. "Optimization of Neural Network for Charpy Toughness of Steel Welds"
   Materials and Manufacturing Processes 24 (2009) 16-21.
   J. H. Pak, J. H. Jang, H. K. D. H. Bhadeshia and L. Karlsson
2. D.J.C. MacKay, 1997, Mathematical Modelling of Weld Phenomena 3, eds. H. Cerjak and H.K.D.H. Bhadeshia, Inst. of Materials, London, pp 359

Top | Next | Prev

Parameters

Program inputs

C Si Mn S P Ni Cr Mo V Cu Co W : wt%

O Ti N B Nb : parts per million by weight

HI(Heat Input) : kJ mm^-1

IT(Interpass temperature) pwhtT(post weld heat treatment Temperature) TT(Test Temperature) : ^oC

pwhtt (post weld heat treatment time) : h

Typical Inputs and corresponding outputs

0.032 0.25 2.02 0.008 0.011 7.23 0.47 0.63 0 0.03 0 0 380 0 250 0 0 1.2 250 250 16 -100
0.032 0.25 2.02 0.008 0.011 7.23 0.47 0.63 0 0.03 0 0 380 0 250 0 0 1.2 250 250 16 -90
0.032 0.25 2.02 0.008 0.011 7.23 0.47 0.63 0 0.03 0 0 380 0 250 0 0 1.2 250 250 16 -80

Outputs: Charpy / J, error / J, Charpy-error, Charpy+error

36.044079 41.092983 -5.048912 77.137062
37.350754 41.509705 -4.158950 78.860458
38.958363 42.062969 -3.104607 81.021332

Top | Next | Prev

Auxiliary Routines

None.

Top | Next | Prev

Keywords

neural network, steel, weld metal, Charpy, toughness

Top | Next | Prev

Download

Download source code (PC)

Download source code, (Unix)

Top | Prev