Artificial Neural Network Modeling for Prediction of Tensile Strength in Pressure Vessel Welding with Steel ASTM A537 Class 1

Journal article

Authors/Editors

Strategic Research Themes

No matching items found.

Publication Details

Author list: ปรัชญา เพียสุระ

Publication year: 2019

Volume number: 7

Issue number: 2

Start page: 36

End page: 49

eISSN: 2672-9989

URL: https://ph01.tci-thaijo.org/index.php/TNIJournal/article/view/206174

Abstract

This research described to the determine a mathematic model using artificial neural network (ANN) for predicting the of tensile strength in the pressure vessel steel ASTM A537 Class1 grade with submerge arc welding process. The following welding parameters were studied: the welding current, voltage and travel speed. The resulting welding samples were examined using tensile strength tests which were observed microstructure with scanning electron microscopy (SEM) and determine a suitable mathematic model. The research results reveal that using a neural network model with the proposed mathematical model, which tensile strength represents 3 neurons for the input 14 neurons and 1 output neurons (3-14-1). The Levenberg-Marquart training algorithm was also train for weight and bias network. The neuron of log-sigmoid for input layer, tan-sigmoid for hidden layer and purelin for output layer activation function was assigned. The mean square error (MSE) and coefficient of determination (R2) for tensile strength predict was showed that of 0.284. The welding conditions which have the highest tensile strength was showed that microstructure phenomenon could be that fine pearlite and spheroidized pearlite with distribute intensity in heat affected zone. The mathematical model that can be effectively applied to predict and quality control of welds to obtain tensile strength according to the standard requirements, which was reduce production costs and increase work efficiency.

Keywords

การเชื่อมใต้ฟลักซ์, ค่าความเค้นแรงดึง, แบบจำลอง, ภาชนะแรงดันสูง, วิธีโครงข่ายประสาทเทียม