쾌삭 303계 스테인리스강 소형 압연 선재 제조 공정의 생산품질 예측 모형
Quality Prediction Model for Manufacturing Process of Free-Machining 303-series Stainless Steel Small Rolling Wire Rods
0
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
This article suggests the machine learning model, i.e., classifier, for predicting the production quality of free-machining 303-series stainless steel(STS303) small rolling wire rods according to the operating condition of the manufacturing process. For the development of the classifier, manufacturing data for 37 operating variables were collected from the manufacturing execution system(MES) of Company S, and the 12 types of derived variables were generated based on literature review and interviews with field experts. This research was performed with data preprocessing, exploratory data analysis, feature selection, machine learning modeling, and the evaluation of alternative models. In the preprocessing stage, missing values and outliers are removed, and oversampling using SMOTE(Synthetic oversampling technique) to resolve data imbalance. Features are selected by variable importance of LASSO(Least absolute shrinkage and selection operator) regression, extreme gradient boosting(XGBoost), and random forest models. Finally, logistic regression, support vector machine(SVM), random forest, and XGBoost are developed as a classifier to predict the adequate or defective products with new operating conditions. The optimal hyper-parameters for each model are investigated by the grid search and random search methods based on -fold cross-validation. As a result of the experiment, XGBoost showed relatively high predictive performance compared to other models with an accuracy of 0.9929, specificity of 0.9372, -score of 0.9963, and logarithmic loss of 0.0209. The classifier developed in this study is expected to improve productivity by enabling effective management of the manufacturing process for the STS303 small rolling wire rods.
Abstract
This article suggests the machine learning model, i.e., classifier, for predicting the production quality of free-machining 303-series stainless steel(STS303) small rolling wire rods according to the operating condition of the manufacturing process. For the development of the classifier, manufacturing data for 37 operating variables were collected from the manufacturing execution system(MES) of Company S, and the 12 types of derived variables were generated based on literature review and interviews with field experts. This research was performed with data preprocessing, exploratory data analysis, feature selection, machine learning modeling, and the evaluation of alternative models. In the preprocessing stage, missing values and outliers are removed, and oversampling using SMOTE(Synthetic oversampling technique) to resolve data imbalance. Features are selected by variable importance of LASSO(Least absolute shrinkage and selection operator) regression, extreme gradient boosting(XGBoost), and random forest models. Finally, logistic regression, support vector machine(SVM), random forest, and XGBoost are developed as a classifier to predict the adequate or defective products with new operating conditions. The optimal hyper-parameters for each model are investigated by the grid search and random search methods based on -fold cross-validation. As a result of the experiment, XGBoost showed relatively high predictive performance compared to other models with an accuracy of 0.9929, specificity of 0.9372, -score of 0.9963, and logarithmic loss of 0.0209. The classifier developed in this study is expected to improve productivity by enabling effective management of the manufacturing process for the STS303 small rolling wire rods.
참고문헌 (24)
- Ahn, Y. S., "Effect of Ti, B, Element and δ-ferrite on the Hot Ductility of STS 304", Korean Journal of Metals and Materials, 30(7), pp.799~806 (1992)
- Ahn, Y, "Effect of Ti, B.Element and δ-ferrite on the Hot Ductility of STS304", Journal of the Korean Inst. of Met. & Mater, 30(7), pp.799~806 (1992)
- Bae, C. M., "Effects of Sulfur and Boron Content on Hot Ductility and Machinability of STS303", Korean Journal of Metals and Materials, 32(6), pp.706~713 (1994)
- Brinkman, C, "Properties of austenitic stainless steels and their weld metals (influence of slight chemistry variations)" (1979), ASTM International
- Cheruvathur, S., "Additive manufacturing of 17-4 PH stainless steel : post-processing heat treatment to achieve uniform reproducible microstructure", JOM, 68(3), pp.930~942 (2016)
- Díaz-Uriarte, R., "Gene selection and classification of microarray data using random forest", BMC Bioinformatics, 7(1), pp.1~13 (2006)
- Jeong, Y., "Defect Type Prediction in Manufacturing Process using Association Rules", KSIE Spring Conference, pp.310~313 (2004)
- 정재영, "쾌삭 303계 스테인레스강의 열간성형성에 미치는 균질화처리, Mn", S 비 및 델타-페라이트의 영향, 2017, pp.55(12), 880~887[DOI]
- Kang, D, "Predicting the quality of steelplate products using artificial neural networks", KSQM Autumn Conference, pp.146~ (2019)
- 김기현, "셀 레벨에서의 OPTICS 기반 특질 추출을 이용한 칩 품질 예측", 대한산업공학회지, 40(3), pp.257~266 (2014), 대한산업공학회[DOI]
- 김상원, "3 wt.% Cu 함유 STS 304 빌렛의 열간가공성과 표면결함에 미치는 δ-ferrite의 영향", 한국재료학회지, 14(6), pp.379~388 (2004), 한국재료학회
- Kim, Y. S., "Review of Formability and Forming Property for Stainless Steel", Transactions of Materials Processing, 20(3), pp.193~205 (2011)
- Lee, S., "Effect of S and Mn on the Hot Workability of STS 316L and 309S steels", Journal of the Korean Inst. of Met. & Mater, 36(10), pp.1590~1598 (1998)
- Lee, Y., "Effect of castring structure and S contents on the how workability of 18-8stainless steel", Korean Journal of Metal and Materials, 1995(1), pp.204~ (1995)
- Liu, H., "Analysis of MnS inclusions formation in resulphurised steel via modeling and experiments", Materials, 12(2), pp.2028~ (2019)
- Maciejewski, J., "The effects of sulfide inclusions on mechanical properties and failures of steel components", Journal of Failure Analysis and Prevention, 15(2), pp.169~178 (2015)
- Nazábal, J. L., "Hightemperature deformation characteristics of free-machining steels", Metals Technology, 9(1), pp.323~326 (1982)
- Nkonyana, T, "Performance evaluation of data mining techniques in steel manufacturing industry", Procedia Manufacturing, 35, pp.623~628
- Oliver, J., "Method for manufacturing and utilizing ferritic-austenitic stainless steel with high formability", U. S. Patent Application, ~ (2013)
- Osakada, K., "Effects of strain rate and temperature in forming processes of metals", Le Journal de Physique IV, 7(C3), pp.C3~XXXVII (1997)
- 박종현, "머신러닝을 활용한 소형 압연 공정 온도 예측 신뢰성 평가", 제어.로봇.시스템학회 논문지, 26(6), pp.412~422 (2020), 제어·로봇·시스템학회[DOI]
- Ruiz, E., "Machine learning algorithms for the prediction of the strength of steel rods : an example of data-driven manufacturing in steelmaking", International Journal of Computer Integrated Manufacturing, 33(9), pp.880~894 (2020)
- 사공운, "스마트 팩토리를 위한 하둡 에코 시스템 및 머신러닝 기반의 고무 공정 데이터 분석", 정보과학회 컴퓨팅의 실제 논문지, 26(12), pp.519~527 (2020), 한국정보과학회[DOI]
- Toledo, G. A., "Influence of sulfur and Mn", S ratio on the hot ductility of steels during continuous casting, 1993, pp.64(6), 292~299