Application of Machine Learning Algorithms and SHAP for Prediction and Feature Analysis of Tempered Martensite Hardness in Low-Alloy Steels
The tempering of low-alloy steels is important for controlling the mechanical properties required for industrial fields. Several studies have investigated the relationships between the input and target values of materials using machine learning algorithms. The limitation of machine learning algorith...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-07-01
|
Series: | Metals |
Subjects: | |
Online Access: | https://www.mdpi.com/2075-4701/11/8/1159 |
id |
doaj-d19427ed037f4cd9968ac3f3750430d5 |
---|---|
record_format |
Article |
spelling |
doaj-d19427ed037f4cd9968ac3f3750430d52021-08-26T14:04:04ZengMDPI AGMetals2075-47012021-07-01111159115910.3390/met11081159Application of Machine Learning Algorithms and SHAP for Prediction and Feature Analysis of Tempered Martensite Hardness in Low-Alloy SteelsJunhyub Jeon0Namhyuk Seo1Seung Bae Son2Seok-Jae Lee3Minsu Jung4Division of Advanced Materials Engineering, Jeonbuk National University, Jeonju 54896, KoreaDivision of Advanced Materials Engineering, Jeonbuk National University, Jeonju 54896, KoreaDivision of Advanced Materials Engineering, Jeonbuk National University, Jeonju 54896, KoreaDivision of Advanced Materials Engineering, Jeonbuk National University, Jeonju 54896, KoreaIntelligent Manufacturing R&D Department, Korea Institute of Industrial Technology, Siheung 15014, KoreaThe tempering of low-alloy steels is important for controlling the mechanical properties required for industrial fields. Several studies have investigated the relationships between the input and target values of materials using machine learning algorithms. The limitation of machine learning algorithms is that the mechanism of how the input values affect the output has yet to be confirmed despite numerous case studies. To address this issue, we trained four machine learning algorithms to control the hardness of low-alloy steels under various tempering conditions. The models were trained using the tempering temperature, holding time, and composition of the alloy as the inputs. The input data were drawn from a database of more than 1900 experimental datasets for low-alloy steels created from the relevant literature. We selected the random forest regression (RFR) model to analyze its mechanism and the importance of the input values using Shapley additive explanations (SHAP). The prediction accuracy of the RFR for the tempered martensite hardness was better than that of the empirical equation. The tempering temperature is the most important feature for controlling the hardness, followed by the C content, the holding time, and the Cr, Si, Mn, Mo, and Ni contents.https://www.mdpi.com/2075-4701/11/8/1159low-alloy steelstemperingtempered martensite hardnessmachine learningSHAP |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Junhyub Jeon Namhyuk Seo Seung Bae Son Seok-Jae Lee Minsu Jung |
spellingShingle |
Junhyub Jeon Namhyuk Seo Seung Bae Son Seok-Jae Lee Minsu Jung Application of Machine Learning Algorithms and SHAP for Prediction and Feature Analysis of Tempered Martensite Hardness in Low-Alloy Steels Metals low-alloy steels tempering tempered martensite hardness machine learning SHAP |
author_facet |
Junhyub Jeon Namhyuk Seo Seung Bae Son Seok-Jae Lee Minsu Jung |
author_sort |
Junhyub Jeon |
title |
Application of Machine Learning Algorithms and SHAP for Prediction and Feature Analysis of Tempered Martensite Hardness in Low-Alloy Steels |
title_short |
Application of Machine Learning Algorithms and SHAP for Prediction and Feature Analysis of Tempered Martensite Hardness in Low-Alloy Steels |
title_full |
Application of Machine Learning Algorithms and SHAP for Prediction and Feature Analysis of Tempered Martensite Hardness in Low-Alloy Steels |
title_fullStr |
Application of Machine Learning Algorithms and SHAP for Prediction and Feature Analysis of Tempered Martensite Hardness in Low-Alloy Steels |
title_full_unstemmed |
Application of Machine Learning Algorithms and SHAP for Prediction and Feature Analysis of Tempered Martensite Hardness in Low-Alloy Steels |
title_sort |
application of machine learning algorithms and shap for prediction and feature analysis of tempered martensite hardness in low-alloy steels |
publisher |
MDPI AG |
series |
Metals |
issn |
2075-4701 |
publishDate |
2021-07-01 |
description |
The tempering of low-alloy steels is important for controlling the mechanical properties required for industrial fields. Several studies have investigated the relationships between the input and target values of materials using machine learning algorithms. The limitation of machine learning algorithms is that the mechanism of how the input values affect the output has yet to be confirmed despite numerous case studies. To address this issue, we trained four machine learning algorithms to control the hardness of low-alloy steels under various tempering conditions. The models were trained using the tempering temperature, holding time, and composition of the alloy as the inputs. The input data were drawn from a database of more than 1900 experimental datasets for low-alloy steels created from the relevant literature. We selected the random forest regression (RFR) model to analyze its mechanism and the importance of the input values using Shapley additive explanations (SHAP). The prediction accuracy of the RFR for the tempered martensite hardness was better than that of the empirical equation. The tempering temperature is the most important feature for controlling the hardness, followed by the C content, the holding time, and the Cr, Si, Mn, Mo, and Ni contents. |
topic |
low-alloy steels tempering tempered martensite hardness machine learning SHAP |
url |
https://www.mdpi.com/2075-4701/11/8/1159 |
work_keys_str_mv |
AT junhyubjeon applicationofmachinelearningalgorithmsandshapforpredictionandfeatureanalysisoftemperedmartensitehardnessinlowalloysteels AT namhyukseo applicationofmachinelearningalgorithmsandshapforpredictionandfeatureanalysisoftemperedmartensitehardnessinlowalloysteels AT seungbaeson applicationofmachinelearningalgorithmsandshapforpredictionandfeatureanalysisoftemperedmartensitehardnessinlowalloysteels AT seokjaelee applicationofmachinelearningalgorithmsandshapforpredictionandfeatureanalysisoftemperedmartensitehardnessinlowalloysteels AT minsujung applicationofmachinelearningalgorithmsandshapforpredictionandfeatureanalysisoftemperedmartensitehardnessinlowalloysteels |
_version_ |
1721191613881909248 |