Materials Algorithms Project Program Library

Program MAP_AL_HARDNESS for 356 and 319 Al cast alloys

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

Provenance of Source Code

B Prasanna Venkataraman,
Department of Metallurgical Engineering,
PSG College of Technology,
Coimbatore - 641004, India.

M Murugananth,
Visiting Professor,
Department of Metallurgical Engineering,
PSG College of Technology,
Coimbatore - 641004, India.

E-mail: sssananth@gmail.com

Added to MAP: November 2007.

Top | Next

Purpose

Purpose of code - To predict the hardness of 356 and 319 Al cast alloys as a function of chemical composition and heat treatment (artificial aging T6).

Top | Next | Prev

Specification

Complete program.

Top | Next | Prev

Description

Al-Si cast alloys are widely used in a number of commercial applications. The present work focuses on the 356 (Al-Si-Mg) and 319 (Al-Si-Cu and A-Si-Cu-Mg) category of Al-Si alloys in predicting the hardness (Brinell Hardness, 10mm diameter ball; 500 kgf applied load) of these alloys in the as cast, solution treated and artificially aged condition (T6).

This program was developed on a RedHat 9.0 Linux machine. The .tar files contains the following files and sub-directories:

• Read_me.txt - This file contains all the relevant information regarding the model such as, the number of data points used, the input variables used, explanations on how to run the model and get the desired results.
• MINMAX - This file contains the range (mathematical) of the data, its mean and standard deviation of input variables.
• test.dat - This file is an example of how the variables have to be entered and used for predictions.
• run_model.sh - This is the shell command to be run on the terminal. This command takes the input from the test.dat and runs the model for the given input.
• result - This file is the output of the test.dat after running the shell command run_model.sh.
• no_of_rows.dat - This file contains the number of rows of data used.
• 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 for each model), _R* (Files containing values for the noise, test error and log predictive error for each model).
• Subdirectory D - outran.x (A normalized output file which was created during the building of the model).
• Subdirectory outprdt - com.dat (The normalized output file containing the committee results).

Top | Next | Prev

References

1. M Tash, F H Samuel, F Mucciardi and H W Doty, 2007, Materials Science and Engineering A, 443, pp. 185 – 201.

Top | Next | Prev

Parameters

Input parameters

The details regarding the Input variables are provided in the Read_me.txt file in the .tar file. The MINMAX file gives the range of the variables used for making the model. When the input is within this range of the input variables, the uncertainty involved with the predictions will be lower or otherwise.

Output parameters

The output is the Hardness (Brinell Hardness) of the respective data provided. The output is written into the result file. The output has the predicted value, error, predicted-error (lower limit) and predicted+error (upper limit).

Top | Next | Prev

Error Indicators

None.

Top | Next | Prev

Accuracy

The uncertainty in predictions is given as a predicted-error and predicted+error.

Top | Next | Prev

Further Comments

None.

Top | Next | Prev

Example

1. Program text

Complete program.

2. Input data

    6.79   0.876   0.035   0.3 0.017   0.1 0   0.0399  0   0   0   (composition for as-cast)
    6.79   0.876   0.035   0.3 0.017   0.1 0   0.0399  540 0   0   (solution treated only)
    6.79   0.876   0.035   0.3 0.017   0.1 0   0.0399  540 200 4   (solution treated and artificially aged)
    
   Each number correspond to a particular input variable and the details of the input variables are given in the Read_me.txt file.

3. Program execution

In the command prompt, type sh run_model.sh

4. Program results

Predicted      Error       Predicted-Error     Predicted+Error
    
62.5016625     1.6713286   60.8303339      64.1729911  (predictions for as-cast composition)    
71.8553768     2.5384887   69.3168881      74.3938655  (for solution treated composition)
98.0564042     1.1411630   96.9152412      99.1975672  (for artificially aged composition)
    

Top | Next | Prev

Auxiliary Routines

None.

Top | Next | Prev

Keywords

Artificial Neural Networks, 356 and 319 Al cast alloys, Hardness (Brinell Hardness).

Top | Next | Prev

Download

Download model for LINUX

Top | Prev