Korean Chemical Engineering Research

The Korean Institute of Chemical Engineers (한국화학공학회)
권호 표지
DOI QR코드

DOI QR Code

Data-driven Modeling for Valve Size and Type Prediction Using Machine Learning

머신 러닝을 이용한 밸브 사이즈 및 종류 예측 모델 개발

  • Chanho Kim (Department of Chemical and Biomolecular Engineering, Yonsei University) ;
  • Minshick Choi (Green Materials and Processes R&D Group, Korea Institute of Industrial Technology) ;
  • Chonghyo Joo (Department of Chemical and Biomolecular Engineering, Yonsei University) ;
  • A-Reum Lee (Samsung E&A Co., Ltd.) ;
  • Yun Gun (Samsung E&A Co., Ltd.) ;
  • Sungho Cho (Samsung E&A Co., Ltd.) ;
  • Junghwan Kim (Department of Chemical and Biomolecular Engineering, Yonsei University)
  • 김찬호 (연세대학교 화공생명공학과) ;
  • 최민식 (한국생산기술연구원 친환경재료공정연구그룹) ;
  • 주종효 (연세대학교 화공생명공학과) ;
  • 이아름 (삼성 E&A) ;
  • 윤건 (삼성 E&A) ;
  • 조성호 (삼성 E&A) ;
  • 김정환 (연세대학교 화공생명공학과)
  • Received : 2023.12.07
  • Accepted : 2024.05.22
  • Published : 2024.08.01
Download PDF
⟨ Previous Next ⟩

Abstract

Valves play an essential role in a chemical plant such as regulating fluid flow and pressure. Therefore, optimal selection of the valve size and type is essential task. Valve size and type have been selected based on theoretical formulas about calculating valve sizing coefficient (Cv). However, this approach has limitations such as requiring expert knowledge and consuming substantial time and costs. Herein, this study developed a model for predicting valve sizes and types using machine learning. We developed models using four algorithms: ANN, Random Forest, XGBoost, and Catboost and model performances were evaluated using NRMSE & R2 score for size prediction and F1 score for type prediction. Additionally, a case study was conducted to explore the impact of phases on valve selection, using four datasets: total fluids, liquids, gases, and steam. As a result of the study, for valve size prediction, total fluid, liquid, and gas dataset demonstrated the best performance with Catboost (Based on R2, total: 0.99216, liquid: 0.98602, gas: 0.99300. Based on NRMSE, total: 0.04072, liquid: 0.04886, gas: 0.03619) and steam dataset showed the best performance with RandomForest (R2: 0.99028, NRMSE: 0.03493). For valve type prediction, Catboost outperformed all datasets with the highest F1 scores (total: 0.95766, liquids: 0.96264, gases: 0.95770, steam: 1.0000). In Engineering Procurement Construction industry, the proposed fluid-specific machine learning-based model is expected to guide the selection of suitable valves based on given process conditions and facilitate faster decision-making.

밸브는 유량과 압력 조절 등의 중요한 역할을 수행하며, 적절한 밸브 사이즈와 유형 선택이 필요하다. Engineering Procurement Construction (EPC) 산업에선 밸브 사이즈 계수(Cv)의 수식적 계산을 바탕으로 사이즈와 유형을 선정해왔으나 이러한 방식은 전문가의 많은 시간과 비용이 요구되어 비효율적이다. 본 연구는 이를 해결하기위해 머신 러닝기법을 이용한 밸브 사이즈 및 유형 예측 모델을 개발하였다. Artificial neural network (ANN), Random Forest, XGBoost, Catboost의알고리즘을 적용하여 모델들을 개발하였으며, 평가 지표로는 사이즈 예측에는 Normalized root mean squared error (NRMSE)와 R2를, 종류 예측에는 F1 score를 적용하였다. 또한, 유체 상에 따른 영향을 확인하고자 유체 전체, 액체, 기체, 스팀의 4개의 데이터 세트로 사례 연구를 진행하였다. 연구 결과, 사이즈의 경우 전체, 액체, 기체에선 Catboost(R2기준, 전체: 0.99216, 액체: 0.98602, 기체: 0.99300. NRMSE 기준, 전체: 0.04072, 액체: 0.04886, 기체: 0.03619)가, 스팀에선 Random Forest가(R2: 0.99028, NRMSE: 0.03493) 가장 뛰어난 모델임을 확인하였다. 종류의 경우 Catboost가 모든 데이터에서 가장 높은 성과를 제시하였다(F1 score 기준, 전체: 0.95766, 액체: 0.96264, 기체: 0.95770, 스팀: 1.0000). 본 연구에서 제안한 모델들을 적용할 경우, 주어진 조건에 따른 밸브 선택을 도와 의사결정 속도를 높여줄 것으로 기대된다.

Keywords

Acknowledgement

본 논문은 "AI 지능화기반 엔지니어링 예측 모델 개발(2023-11-0458)"의 지원으로 수행한 연구입니다.

References

  1. Park, G., "How to Select Control Valve," HWAHAK KONGHAK, 10(3), 141-152(1972).
  2. Driskell, Les. Control valve selection and sizing. 1st ed. North Carolina: Creative Services Inc; 1983.
  3. IEC 60534-2-1 Mod: flow equations for sizing control valves, Switzerland, International Electro-technical Commission.
  4. ISA75.01, Control Valve Sizing Equations, International Society of Automation.
  5. Grace, A. and Frawley, P., "Experimental Parametric Equation for the Prediction of Valve Coefficient (Cv) for Choke Valve Trims," International Journal of Pressure Vessels and Piping, 88(2-3), 109-118(2011).
  6. Long, C. and Guan, J., "A Method for Determining Valve Coefficient and Resistance Coefficient for Predicting Gas Flowrate," Experimental Thermal and Fluid Science, 35(6), 1162-1168(2011).
  7. Zhou, X.-M., Wang, Z.-K., Zhang, Y.-F., "A Simple Method for High-precision Evaluation of Valve Flow Coefficient by Computational Fluid Dynamics Simulation," Advances in Mechanical Engineering, 9(7), 1-7(2017).
  8. Lisowski, E. and Filo, G., "Analysis of a Proportional Control Valve Flow Coefficient with the Usage of a CFD Method," 53, Part B, 269-278(2017).
  9. Valdes, J. R., Rodriguez, J. M., Saumell, J, Putz, T., "A Methodology for the Parametric Modelling of the Flow Coefficients and Flow Rate in Hydraulic Valves," Energy Conversion and Management, 88, 598-611(2014).
  10. Nguyen, Q. K. and Jung, K. H., "Experimental Study on Pressure Characteristics and Flow Coefficient of Butterfly Valve," International Journal of Naval Architecture and Ocean Engineering, 15, (2023).
  11. Al-Zaidi, B. M. and Ismaeel, A. J., "Effect of Hydraulic Characteristics on Fluid Transients Analysis under Different Types of Control Valves," Journal of Ecological Engineering, 23(12), 111-123(2022).
  12. Fu, W.-S. and Ger, J.-S., "A Concise Method for Determining a Valve Flow Coefficient of a Valve Under Compressible Gas Flow," Experimental Thermal and Fluid Science, 18, 307-313(1999).
  13. Boccardi, G., Bubbico, R. and Celata, G. P., "Geometry Influence on Safety Valves Sizing in Two-phase Flow," Journal of Loss Prevention in the Process Industries, 21(1), 66-73(2008).
  14. Mahalleh, VBS, YOLO-Based Valve Type Recognition and Localization, 2019 IEEE 6th International Conference On Industrial Engineering and Applications (Iciea), 37-40(2019).
  15. Hlubek, N. and Baumann, M., Sebastian Heinze Florian Ostermaier, Using Machine Learning for Diaphragm Prediction in Solenoid Valves, IEEE 27th International Conference on Emerging Technologies and Factory Automation (ETFA).
  16. Roh, J., Park, H., Kwon, H., Joo, C., Moon, I., Cho, H., Ro, I. and Kim, J., "Interpretable Machine Learning Framework for Catalyst Performance Prediction and Validation with Dry Reforming of Methane," Applied Catalysis B: Environmental, 343, 123454(2024).
  17. Roh, J., Oh, S., Lee, D., Joo, C., Park, J., Moon, I., Ro, I., Kim, J., "Hybrid Quantum Neural Network Model with Catalyst Experimental Validation: Application for the Dry Reforming of Methane," ACS Sustainable Chemistry & Engineering, 12(10), 4121-4131(2024).
  18. Kwon, H., Oh, K. C., Choi, Y., Chung, Y. G. and Kim, J., "Development and Application of Machine Learning-based Prediction Model for Distillation Column," International Journal of Intelligent Systems, 36(5), 1970-1997(2021).
  19. Jeong, S., Joo, C., Lim, J., Cho, H., Lim, S. and Kim, J., "A Novel Graph-based Missing Values Imputation Method for Industrial Lubricant Data," Computers in Industry, 150, 103937(2023).
  20. Lee, J., Hong, S., Kim, J. and Moon, I., "Machine Learning-based Energy Optimization for on-site SMR Hydrogen Production," Energy Conversion and Management, 244(15), 114438(2021).
  21. Lim, J., Jeong, S. and Kim, J., "Deep Neural Network-based Optimal Selection and Blending Ratio of Waste Seashells as an Alternative to High-grade Limestone Depletion for SOX Capture and Utilization," Chemical Engineering Journal, 431, Part 3, 133244(2022).
  22. Joo, C., Park, H. and Kim, J., "Development of Physical Property Prediction Models for Polypropylene Composites with Optimizing Random Forest Hyperparameters," International Journal of Intelligent Systems, 37(6), 3189-3771(2022).
  23. Joo, C., Park, H., Kwon, H. and Kim, J., "Machine Learning Approach to Predict Physical Properties of Polypropylene Composites: Application of MLR, DNN, and Random Forest to Industrial Data," Polymers, 14(17), 3500(2022).
  24. Joo, C., Park, H. and Kim, J., "Data-driven Modeling for Physical Property Prediction of Polypropylene Composites Using Artificial Neural Network and Principal Component Analysis," Computer Aided Chemical Engineering, 51, 1369-1374(2022).
  25. Lee, Y., Choi, Y., Cho, H. and Kim, J., "Prediction of Distillation Column Temperature Using Machine Learning and Data Preprocessing," Korean Chem. Eng. Res., 59(2), 191-199(2021).
  26. Joo, C., Park, H., Lim, J., Cho, H. and Kim, J., "Machine Learning-based Heat Deflection Temperature Prediction and Effect Analysis in Polypropylene Composites Using Catboost and Shapley Additive Explanations," Engineering Applications of Artificial Intelligence, 126, Part A, 1801-1806(2022).
  27. Chen, T. and Guestrin, C., XGBoost: A Scalable Tree Boosting System, 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD), 785-794(2016).
  28. Dorogush, A. V., Ershov, V. and Gulin, A., "CatBoost: Gradient Boosting with Categorical Features Support," Workshop on ML Systems at NIPS 2017.