Open AccessEstimation of Austenitizing and Multiple Tempering Temperatures from the Mechanical Properties of AISI 410 using Artificial Neural Network
Author(s) -
Abdul Kareem F. Hassan,
Qahtan A. Jawad
Publication year - 2018
Publication title -
international journal of engineering and technology
Language(s) - English
Resource type - Journals
ISSN - 2227-524X
DOI - 10.14419/ijet.v7i4.19.27997
Subject(s) - tempering , materials science , ultimate tensile strength , quenching (fluorescence) , austenite , artificial neural network , hardening (computing) , martensite , metallurgy , nonlinear regression , linear regression , izod impact strength test , regression analysis , composite material , machine learning , microstructure , computer science , physics , layer (electronics) , quantum mechanics , fluorescence
This research involved a study of the heat treatment conditions effect on the mechanical properties of martensitic stainless steel type AISI 410. Heat treatment process was hardening of the metal by quenching at different temperature 900°C, 950°C, 1000°C, 1050°C and 1100°C, followed by double tempering at 200°C, 250°C, 300°C, 350°C, 400°C, 450°C, 500°C, 550°C, 600°C, 650°C and 700°C, were evaluated and study of some mechanical properties such as hardness, impact energy and properties of tensile test such as yield and tensile strength is carried out. Multiple outputs Artificial Neural Network model was built with a Matlab package to predict the quenching and tempering temperatures. Also, linear and nonlinear regression analyses (using Data fit package) were used to estimate the mathematical relationship between quenching and tempering temperatures with hardness, impact energy, yield, and tensile strength. A comparison between experimental, regression analysis and ANN model show that the multiple outputs ANN model is more accurate and closer to the experimental results than the regression analysis results.