Open Access

ARTICLE

Predicting Grain Orientations of 316 Stainless Steel Using Convolutional Neural Networks

Dhia K. Suker, Ahmed R. Abdo^*, Khalid Abdulkhaliq M. Alharbi

Mechanical Engineering Department, College of Engineering and Architecture, Umm Al-Qura University, Makkah, 21955, Saudi Arabia

* Corresponding Author: Ahmed R. Abdo. Email:

Intelligent Automation & Soft Computing 2024, 39(5), 929-947. https://doi.org/10.32604/iasc.2024.056341

Received 20 July 2025; Accepted 13 September 2025; Issue published 31 October 2024

Abstract

This paper presents a deep learning Convolutional Neural Network (CNN) for predicting grain orientations from electron backscatter diffraction (EBSD) patterns. The proposed model consists of multiple neural network layers and has been trained on a dataset of EBSD patterns obtained from stainless steel 316 (SS316). Grain orientation changes when considering the effects of temperature and strain rate on material deformation. The deep learning CNN predicts material orientation using the EBSD method to address this challenge. The accuracy of this approach is evaluated by comparing the predicted crystal orientation with the actual orientation under different conditions, using the Root-Mean-Square Error (RMSE) as the measure. Results show that changing the temperature causes different grain orientations to form, meeting the requirements. Further investigations were conducted to validate the results.

---

Keywords

---