Neural Network-Based Evaluation of Hardness in Cold-Rolled Austenitic Stainless Steel Under Various Heat Treatment Conditions
Source abstract
Abstract
This study introduces an innovative, non-contact method for classifying the hardness of austenitic stainless steels (grade AISI 304) based on their intrinsic magnetic fields. Utilizing a 3 × 3 matrix sensor system, this research captures weak magnetic fields to produce precise 2D magnetic field maps of the samples. A key advancement is the application of a modified GoogleNet convolutional neural network, optimized with the stochastic gradient descent with momentum algorithm, which achieves exceptional classification accuracy, ranging from 95% to 100%, and median accuracies of 97.5% to 99%. This method stands out by revealing a novel correlation between annealing temperature and magnetic field strength, particularly a pronounced decline in magnetic properties at temperatures near 1000 °C. This observation underscores the sensitivity of magnetic profiles to heat treatments, offering a groundbreaking approach to material characterization. By enabling reliable, efficient, and fully automated hardness evaluation based on magnetic signatures, this work has the potential to transform materials engineering and manufacturing, setting a new benchmark for non-destructive material analysis techniques.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Microstructure and Mechanical Properties of Steels · Welding Techniques and Residual Stresses · Metal Alloys Wear and Properties
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Electromagnetic and Biomedical Engineering, Faculty of Electrical Engineering and Information Technology, University of Zilina, Univerzitna 8215/1, 010 26 Zilina, Slovakia
Secondary Vocational School of Electrical Engineering, Komenskeho 50, 010 01 Zilina, Slovakia
Department of Mechatronics and Electronics, Faculty of Electrical Engineering and Information Technology, University of Zilina, Univerzitna 8215/1, 010 26 Zilina, Slovakia
References (28)
Show source-provided references
1. Electromagnetic non-destructive evaluation: Present and future10.5545/sv-jme.2010.171 ↗
2. Uncertainty metric in model-based eddy current inversion using the adaptive Monte Carlo method10.1016/j.measurement.2019.01.004 ↗
3. Fast simulation of ECT signal due to a conductive crack of arbitrary width10.1109/TMAG.2006.870973 ↗
4. Rao, B.P.C. (2007). Practical Eddy Current Testing, Alpha industries International Ltd.
5. Rotating field eddy current probe with bobbin pickup coil for steam generator tubes inspection10.1016/j.ndteint.2012.12.002 ↗
6. Balageas, D., Fritzen, C.P., and Güemes, A. (2006). Structural Health Monitoring, ISTE Ltd.. [1st ed.].10.1002/9780470612071 ↗
7. Strapacova, T., Smetana, M., and Capova, K. (2014). Material defect identification using sweep frequency eddy current technique. Studies Electromagnetic Nondestructive Evaluation (XVII), IOS Press.
8. Non-destructive testing of conductive material by eddy current air probe based on swept frequency
9. Microstructural Evolutions During Annealing of Plastically Deformed AISI 304 Austenitic Stainless Steel: Martensite Reversion, Grain Refinement, Recrystallization, and Grain Growth10.1007/s11661-016-3589-1 ↗
10. Effect of Heat Treatment Temperature on Mechanical Properties of the AISI 304 Stainless Steel
11. Collings, E.W. (1980). The Magnetic Character of Austenitic Stainless Steels. Advances in Cryogenic Engineering Materials, Springer.10.1007/978-1-4613-9859-2_3 ↗
12. Change of magnetic properties in austenitic stainless steels due to plastic deformation10.1016/j.prostr.2018.12.352 ↗
13. Solution of Inverse Problems in Electromagnetic NDT Using Neural Networks
14. Fast numerical techniques for electromagnetic nondestructive evaluation10.1080/10589750802195568 ↗
15. Szymanik, B., Hashemi, M., Lopato, P., and Psuj, G. (2021). Detection and Identification of Defects in 3D-Printed Dielectric Structures via Thermographic Inspection and Deep Neural Networks. Materials, 14.10.3390/ma14154168 ↗
16. Micromagnetic and Quantitative Prediction of Surface Hardness in Carbon Steels Based on a Joint Classification-Regression Method10.1007/s10921-022-00893-y ↗
17. Xiao, W., Xiong, J., Yu, L., He, Y., and Ma, G. (2023). Weld Defect Monitoring Based on Two-Stage Convolutional Neural Network. J. Shanghai Jiaotong Univ., 1–9.10.1007/s12204-023-2608-9 ↗
18. Intelligent Hardness Prediction of Bearing Rings Using Pulsed Eddy Current Testing10.1109/JSEN.2022.3212468 ↗
19. Habeeb, M., You, H.W., Aher, K.B., Bhavar, G.B., Pawar, S.S., and Gaikwad, S.D. (2023). Artificial neural networks for the prediction of mechanical properties of CGNP/PLGA nanocomposites. Mater. Today Proc., in press.10.1016/j.matpr.2023.08.354 ↗
20. Evaluation of Defects Depth for Metal Sheets Using Four-Coil Excitation Array Eddy Current Sensor and Improved ResNet18 Network10.1109/JSEN.2024.3367816 ↗
21. Hassani, S., and Dackermann, U. (2023). A Systematic Review of Advanced Sensor Technologies for Non-Destructive Testing and Structural Health Monitoring. Sensors, 23.10.3390/s23042204 ↗
22. Detection of fatigue degradation in austenitic stainless steel with eddy current probe and machine learning10.1016/j.jmrt.2023.11.176 ↗
23. Waste material classification using performance evaluation of deep learning models
24. Liu, H., and Lang, B. (2019). Machine Learning and Deep Learning Methods for Intrusion Detection Systems: A Survey. Appl. Sci., 9.10.3390/app9204396 ↗
25. Baldominos, A., Cervantes, A., Saez, Y., and Isasi, P. (2019). A Comparison of Machine Learning and Deep Learning Techniques for Activity Recognition using Mobile Devices. Sensors, 19.10.3390/s19030521 ↗
26. Sensitec (2024, October 12). AFF755B Datasheet. Available online: https://www.sensitec.com.
27. Murphy, K.P. (2012). Machine Learning: A Probabilistic Perspective, MIT Press.
28. (2024, October 15). Mathworks MATLAB. Available online: https://www.mathworks.com/help/deeplearning/ref/nnet.cnn.trainingoptionssgdm.html.