RISS 학술연구정보서비스

검색

인기 검색어

    다국어 입력

    http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.

    변환된 중국어를 복사하여 사용하시면 됩니다.

    예시)
    • 中文 을 입력하시려면 zhongwen을 입력하시고 space를누르시면됩니다.
    • 北京 을 입력하시려면 beijing을 입력하시고 space를 누르시면 됩니다.
    닫기

    Robust Adaptation Strategies of Data-Driven Models for Chemical Process System = 화학 공정 시스템을 위한 데이터 기반 모델의 강건한 적응 전략

    한글로보기

    https://www.riss.kr/link?id=T17110062

    • 0

      상세조회
    • 0

      다운로드
    서지정보 열기
    • 내보내기
    • 내책장담기
    • 공유하기
    • 오류접수

    부가정보

    다국어 초록 (Multilingual Abstract) kakao i 다국어 번역

    The modeling of chemical processes is an important step in the simulation, optimization, and control of systems, and various studies have employed both first-principles and data-driven methods. However, these methods face several challenges when modeling industrial processes. First-principles modeling can construct reliable models if the mechanism of the process is accurately understood. However, the increasing complexity of modern processes poses challenges to accurately capturing their mechanics. Data-driven modeling does not require prior knowledge of the process, making it applicable to complex systems. However, in industrial processes, data is often limited to normal operating conditions, which can lead to overfitting of the data-driven model. In addition, processes are subject to rectifications and changes in operating conditions, posing the challenge of acquiring sufficient data for the modified process. Therefore, to apply data-driven modeling in industrial chemical processes, adaptation strategies are needed to address the aforementioned challenges.
    In this regard, this thesis proposes effective and robust adaptive strategies to address the challenges faced during data-driven modeling of industrial chemical processes. First, the application of envelope Bayesian optimization is proposed to minimize the data needed for optimizing the catalyst packing ratio in the Fischer-Tropsch microchannel reactor. The envelope Bayesian optimization was modified to accommodate the different optimization dimensions of single-channel and 4-channel reactors. The proposed optimizer not only reduces the data required to reach the optimal point compared to generic Bayesian optimization but also demonstrates robust optimization performance even when the optimization range varies. Second, a hybrid modeling approach is proposed for predicting concentration, integrating prior process knowledge with data-driven modeling methods. This approach leverages a first-principles model of the process to enhance the accuracy and robustness of concentration prediction. The proposed hybrid model improves generalization performance and exhibits better prediction accuracy compared to the data-driven model when encountering process changes. Moreover, the hybrid model is used to construct a refractive index fault detection model by augmenting the monomer concentration of the stream which has longer sampling frequency intervals. This fault detection model demonstrates higher performance compared to models relying solely on existing process data.
    Proposed robust adaptation strategies to overcome the challenges of implementing data-driven modeling in industrial chemical processes, encompassing issues such as acquiring new process data, handling data imbalance, and addressing differences in sampling frequency. These adaptation strategies were then applied to various chemical processes, demonstrating improved performance compared to conventional modeling methods.
    번역하기

    The modeling of chemical processes is an important step in the simulation, optimization, and control of systems, and various studies have employed both first-principles and data-driven methods. However, these methods face several challenges when model...

    The modeling of chemical processes is an important step in the simulation, optimization, and control of systems, and various studies have employed both first-principles and data-driven methods. However, these methods face several challenges when modeling industrial processes. First-principles modeling can construct reliable models if the mechanism of the process is accurately understood. However, the increasing complexity of modern processes poses challenges to accurately capturing their mechanics. Data-driven modeling does not require prior knowledge of the process, making it applicable to complex systems. However, in industrial processes, data is often limited to normal operating conditions, which can lead to overfitting of the data-driven model. In addition, processes are subject to rectifications and changes in operating conditions, posing the challenge of acquiring sufficient data for the modified process. Therefore, to apply data-driven modeling in industrial chemical processes, adaptation strategies are needed to address the aforementioned challenges.
    In this regard, this thesis proposes effective and robust adaptive strategies to address the challenges faced during data-driven modeling of industrial chemical processes. First, the application of envelope Bayesian optimization is proposed to minimize the data needed for optimizing the catalyst packing ratio in the Fischer-Tropsch microchannel reactor. The envelope Bayesian optimization was modified to accommodate the different optimization dimensions of single-channel and 4-channel reactors. The proposed optimizer not only reduces the data required to reach the optimal point compared to generic Bayesian optimization but also demonstrates robust optimization performance even when the optimization range varies. Second, a hybrid modeling approach is proposed for predicting concentration, integrating prior process knowledge with data-driven modeling methods. This approach leverages a first-principles model of the process to enhance the accuracy and robustness of concentration prediction. The proposed hybrid model improves generalization performance and exhibits better prediction accuracy compared to the data-driven model when encountering process changes. Moreover, the hybrid model is used to construct a refractive index fault detection model by augmenting the monomer concentration of the stream which has longer sampling frequency intervals. This fault detection model demonstrates higher performance compared to models relying solely on existing process data.
    Proposed robust adaptation strategies to overcome the challenges of implementing data-driven modeling in industrial chemical processes, encompassing issues such as acquiring new process data, handling data imbalance, and addressing differences in sampling frequency. These adaptation strategies were then applied to various chemical processes, demonstrating improved performance compared to conventional modeling methods.

    더보기

    국문 초록 (Abstract) kakao i 다국어 번역

    화학 공정의 모델링은 시스템의 시뮬레이션과 최적화 그리고 제어를 하는 데 있어 필수적인 과정이다. 따라서 제일 원리 또는 데이터 기반 방법론을 이용하여 모델링을 하는 연구들이 진행해 왔다. 그러나 실제 산업 공정들을 모델링 하기에 기존의 방법들은 몇 가지 문제점들이 존재한다.
    먼저 제일원리 기반 모델링의 경우 프로세스의 메커니즘이 정확히 파악될 경우 높은 신뢰도의 모델을 얻을 수 있지만, 화학 산업 공정의 경우 복잡한 시스템을 가지고 있어 신뢰도 있는 모델을 구성하는 데 비용과 노력이 많이 든다. 데이터 기반 모델링은 공정에 대한 사전 지식이 필요하지 않아 복잡한 시스템에 적용이 용이하다. 그러나 실제 산업 공정의 경우 잘 제어된 시스템이기 때문에 정상운전 범위 내의 데이터가 주를 이루기 때문에 모델의 과적합 문제를 발생시키게 된다. 그리고 산업 현장에서는 공정의 수정이나 운전 시스템의 변화가 발생하는데, 이 경우 변화된 프로세스의 데이터를 충분히 얻어야 한다는 문제가 있다. 따라서 데이터 기반 모델링을 실제 화학 산업 공정에 적용하기 위해서는 앞서 언급한 문제점들을 해결할 수 있는 적응 전략이 필요하다.
    이러한 관점을 기반으로 본 논문은 실제 화학 산업 공정에 데이터 기반 모델링 과정에서 발생하는 문제점들을 해결하기 위한 효과적이고 강건한 적응전략을 제시한다. 먼저 피셔-트롭쉬 마이크로채널 반응기의 촉매 충전 비율을 최적화하는 데 필요한 데이터의 수를 줄이기 위해 엔벨로프 베이지안 최적화를 적용하는 방법을 제시하였다. 이 때, 단일 채널 반응기와 4-채널 반응기의 최적화 차원이 달라지는 것을 해결하기 위해 엔벨로프 베이지안 최적화 방법론을 수정하여 적용하였다. 제안한 최적화 방법론은 기존 베이지안 최적화에 비해 최적점에 도달하는 데 필요한 데이터 수가 적을 뿐 아니라 최적화 범위가 달라지더라도 강건한 최적화 성능을 보였다.
    둘째로, 제일 원리 모델링과 데이터 기반 모델링을 결합한 하이브리드 모델링 기법을 제안하여 산업 공정의 데이터 불균형 및 샘플링 빈도 차이로 발생하는 문제를 해결하였다. LSTM과 제일 원리 모델을 결합하여 프로세스 내의 단량체 조성을 예측 진행하였다. 제안한 하이브리드 모델을 공정에 대한 사전 지식을 데이터 기반 모델링에 결합함으로써 모델의 일반화 성능을 높여 데이터 기반 모델링 보다 공정의 변화가 생겼을 때의 뛰어난 예측 성능을 보였다. 개발한 하이브리드 모델을 통해 샘플링 빈도 주기가 긴 스트림 내 단량체의 조성 값을 대체하여 굴절률 이상 감지 모델을 개발하였다. 개발한 이상 감지 모델은 기존 데이터만을 쓴 모델에 비해 높은 예측 정확도를 나타내었다.
    본 논문은 실제 화학 산업 공정에 데이터 기반 모델링 적용시 발생하는 문제들인 새로운 프로세스에 대한 데이터 필요성과 데이터 불균형 및 샘플링 빈도 차이를 해결할 수 있는 강건한 적응 전략을 제시하였다. 제시한 적응 전략들을 여러 화학 공정에 적용하여 기존 방법론들 대비 제안한 방법론의 개선된 성능을 보여 주었다.
    번역하기

    화학 공정의 모델링은 시스템의 시뮬레이션과 최적화 그리고 제어를 하는 데 있어 필수적인 과정이다. 따라서 제일 원리 또는 데이터 기반 방법론을 이용하여 모델링을 하는 연구들이 진행...

    화학 공정의 모델링은 시스템의 시뮬레이션과 최적화 그리고 제어를 하는 데 있어 필수적인 과정이다. 따라서 제일 원리 또는 데이터 기반 방법론을 이용하여 모델링을 하는 연구들이 진행해 왔다. 그러나 실제 산업 공정들을 모델링 하기에 기존의 방법들은 몇 가지 문제점들이 존재한다.
    먼저 제일원리 기반 모델링의 경우 프로세스의 메커니즘이 정확히 파악될 경우 높은 신뢰도의 모델을 얻을 수 있지만, 화학 산업 공정의 경우 복잡한 시스템을 가지고 있어 신뢰도 있는 모델을 구성하는 데 비용과 노력이 많이 든다. 데이터 기반 모델링은 공정에 대한 사전 지식이 필요하지 않아 복잡한 시스템에 적용이 용이하다. 그러나 실제 산업 공정의 경우 잘 제어된 시스템이기 때문에 정상운전 범위 내의 데이터가 주를 이루기 때문에 모델의 과적합 문제를 발생시키게 된다. 그리고 산업 현장에서는 공정의 수정이나 운전 시스템의 변화가 발생하는데, 이 경우 변화된 프로세스의 데이터를 충분히 얻어야 한다는 문제가 있다. 따라서 데이터 기반 모델링을 실제 화학 산업 공정에 적용하기 위해서는 앞서 언급한 문제점들을 해결할 수 있는 적응 전략이 필요하다.
    이러한 관점을 기반으로 본 논문은 실제 화학 산업 공정에 데이터 기반 모델링 과정에서 발생하는 문제점들을 해결하기 위한 효과적이고 강건한 적응전략을 제시한다. 먼저 피셔-트롭쉬 마이크로채널 반응기의 촉매 충전 비율을 최적화하는 데 필요한 데이터의 수를 줄이기 위해 엔벨로프 베이지안 최적화를 적용하는 방법을 제시하였다. 이 때, 단일 채널 반응기와 4-채널 반응기의 최적화 차원이 달라지는 것을 해결하기 위해 엔벨로프 베이지안 최적화 방법론을 수정하여 적용하였다. 제안한 최적화 방법론은 기존 베이지안 최적화에 비해 최적점에 도달하는 데 필요한 데이터 수가 적을 뿐 아니라 최적화 범위가 달라지더라도 강건한 최적화 성능을 보였다.
    둘째로, 제일 원리 모델링과 데이터 기반 모델링을 결합한 하이브리드 모델링 기법을 제안하여 산업 공정의 데이터 불균형 및 샘플링 빈도 차이로 발생하는 문제를 해결하였다. LSTM과 제일 원리 모델을 결합하여 프로세스 내의 단량체 조성을 예측 진행하였다. 제안한 하이브리드 모델을 공정에 대한 사전 지식을 데이터 기반 모델링에 결합함으로써 모델의 일반화 성능을 높여 데이터 기반 모델링 보다 공정의 변화가 생겼을 때의 뛰어난 예측 성능을 보였다. 개발한 하이브리드 모델을 통해 샘플링 빈도 주기가 긴 스트림 내 단량체의 조성 값을 대체하여 굴절률 이상 감지 모델을 개발하였다. 개발한 이상 감지 모델은 기존 데이터만을 쓴 모델에 비해 높은 예측 정확도를 나타내었다.
    본 논문은 실제 화학 산업 공정에 데이터 기반 모델링 적용시 발생하는 문제들인 새로운 프로세스에 대한 데이터 필요성과 데이터 불균형 및 샘플링 빈도 차이를 해결할 수 있는 강건한 적응 전략을 제시하였다. 제시한 적응 전략들을 여러 화학 공정에 적용하여 기존 방법론들 대비 제안한 방법론의 개선된 성능을 보여 주었다.

    더보기

    목차 (Table of Contents)

    • Abstract i
    • 1. Introduction 1
    • 1.1 Research motivation 1
    • 1.2 Research objectives 3
    • 1.3 Outline of the thesis 5
    • Abstract i
    • 1. Introduction 1
    • 1.1 Research motivation 1
    • 1.2 Research objectives 3
    • 1.3 Outline of the thesis 5
    • 2. Backgrounds and Preliminaries 7
    • 2.1 Bayesian optimization 7
    • 2.1.1 Gaussian process regression 9
    • 2.1.2 Acquisition function 10
    • 2.2 Errors-in-variables model 11
    • 3. Transfer Learning Approach for Optimization of Catalyst Packing Ratio during Scale-Up of Microchannel Reactor 16
    • 3.1 Introduction 16
    • 3.2 Reactor modeling 20
    • 3.2.1 Fischer-Tropsch reaction kinetics 20
    • 3.2.2 Microchannel reactor model 23
    • 3.2.3 Model geometry 26
    • 3.3 Optimization scheme 31
    • 3.3.1 Formulation of optimization problem 31
    • 3.3.2 Computational fluid dynamics-enveloped Bayesian optimization 33
    • 3.4 Simulation conditions and setup 41
    • 3.4.1 Simulation conditions 41
    • 3.4.2 Simulation setup 42
    • 3.4.3 Optimization setup 44
    • 3.5 Results and discussion 46
    • 3.5.1 CFD model validation using short-channel reactor 46
    • 3.5.2 Optimization of 4-channel reactor using CFD-EBO 52
    • 3.5.3 Changes in catalyst packing ratio range 64
    • 4. Hybrid Modeling Approach for Concentration and Fault Prediction in Terpolymerization Process 72
    • 4.1 Introduction 72
    • 4.2 Process description 75
    • 4.3 Data preprocessing 78
    • 4.4 Proposed hybrid modeling approach 80
    • 4.4.1 First-principles model 82
    • 4.4.2 Data-driven model 89
    • 4.4.3 Simulation using hybrid model 99
    • 4.5 Refractive index fault detection 112
    • 5. Concluding remarks 118
    • 5.1 Conclusions 118
    • 5.2 Future work 119
    • Bibliography 121
    더보기

    참고문헌 (Reference)

    1. Collaborative hyperparameter tuning, M. Brendel, B. Kégl and, R. Bardenet, M. Sebag, International Conference on Machine Learning, pp. 199– 207, OMLR, , 2013

    2. Mechanisms of catalyst deactivation, C. H. Bartholomew, Applied Catalysis A: General, vol. 212, no. 1-2, pp. 17–60, , 2001

    3. Sensitivity analysis for chemical models, S. Tarantola and, A. Saltelli, M. Ratto, F. Campolongo, vol. 105, no. 7, pp. 2811–2828, , 2005

    4. Trends in computer-aided process modeling, W. Marquardt, Computers & Chemical Engineering, vol. 20, no. 6-7, pp. 591–609, , 1996

    5. A bayesian study of the error-in-variables model, P. M. Reilly and, H. Patino-Lea1, Technometrics, vol. 23, no. 3, pp. 221–231, , 1981

    6. Evaluation of the terpolymer composition equation, D. J. Kahn and, H. H. Horowitz, vol. 54, no. 160, pp. 363–374, , 1961

    7. Recent trends on hybrid modeling for industry 4.0, J. Sansana, L. H. Chiang and, Z. Wang, R. Rendall, I. Castillo, M. S. Reis, M. N. Joswiak, vol. 151, p. 107365,, , 2021

    8. Copolymerizationthe composition distribution curve, I. Skeist, vol. 68, no. 9, pp. 1781–1784, , 1946

    9. Google vizierA service for black-box optimization in, S. Moitra, D. Sculley, G. Kochanski, B. Solnik, J. Karro and, D. Golovin, Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 1487–1495, , 2017

    10. A survey of predictive modeling on imbalanced domains, R. P. Ribeiro, P. Branco, L. Torgo and, ACM Computing Surveys, vol. 49, no. 2, pp. 1–50, , 2016

    1. Collaborative hyperparameter tuning, M. Brendel, B. Kégl and, R. Bardenet, M. Sebag, International Conference on Machine Learning, pp. 199– 207, OMLR, , 2013

    2. Mechanisms of catalyst deactivation, C. H. Bartholomew, Applied Catalysis A: General, vol. 212, no. 1-2, pp. 17–60, , 2001

    3. Sensitivity analysis for chemical models, S. Tarantola and, A. Saltelli, M. Ratto, F. Campolongo, vol. 105, no. 7, pp. 2811–2828, , 2005

    4. Trends in computer-aided process modeling, W. Marquardt, Computers & Chemical Engineering, vol. 20, no. 6-7, pp. 591–609, , 1996

    5. A bayesian study of the error-in-variables model, P. M. Reilly and, H. Patino-Lea1, Technometrics, vol. 23, no. 3, pp. 221–231, , 1981

    6. Evaluation of the terpolymer composition equation, D. J. Kahn and, H. H. Horowitz, vol. 54, no. 160, pp. 363–374, , 1961

    7. Recent trends on hybrid modeling for industry 4.0, J. Sansana, L. H. Chiang and, Z. Wang, R. Rendall, I. Castillo, M. S. Reis, M. N. Joswiak, vol. 151, p. 107365,, , 2021

    8. Copolymerizationthe composition distribution curve, I. Skeist, vol. 68, no. 9, pp. 1781–1784, , 1946

    9. Google vizierA service for black-box optimization in, S. Moitra, D. Sculley, G. Kochanski, B. Solnik, J. Karro and, D. Golovin, Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 1487–1495, , 2017

    10. A survey of predictive modeling on imbalanced domains, R. P. Ribeiro, P. Branco, L. Torgo and, ACM Computing Surveys, vol. 49, no. 2, pp. 1–50, , 2016

    11. Chemical kinetics and modeling of combustion processes, F. L. Dryer, C. K. Westbrook and, Symposium (International) on Combustion, vol. 18, pp. 749–767, Elsevier, , 1981

    12. Copolymerization of systems containing three components, G. Goldfinger, T. Alfrey and, vol. 14, no. 2, pp. 115–116, , 1946

    13. Deep double descentWhere bigger models and more data hurt, P. Nakkiran, G. Kaplun, B. Barak and, Y. Bansal, T. Yang, I. Sutskever, vol. 2021, no. 12, p. 124003,, , 2021

    14. Properties and morphologies of pvc/nylon terpolymer blends, R. Fan and, Y. Zhang, Y. Lian, X. Zhang, Y. Zhang, Z. Peng, vol. 80, no. 14, pp. 2823–2832, , 2001

    15. A parameter estimation program for the error-invariable model, S. Tang and, Y. Wang, vol. 156, no. 2-3, pp. 225–236, , 2002

    16. Acrylonitrile polymers survey and styrene-acrylonitrile (san), M. M. Wu, Kirk-Othmer Encyclopedia of Chemical Technology, , 2000

    17. A data-based process modeling approach and its applications in, S. J. Qin and, T. J. McAvoy, Dynamics and Control of Chemical Reactors, Distillation Columns and Batch Processes, pp. 93–98, Elsevier, , 1993

    18. Representation of terpolymer kinetics on triangular coordinates, D. Cruise and, R. Lacombe, Journal of Polymer Science Part A-1: Polymer Chemistry, vol. 8, no. 6, pp. 1373–1383, , 1970

    19. Application of refractive index measurements to polymer analysis, R. J. Samuels, vol. 26, no. 4, pp. 1383– 1412, , 1981

    20. Data augmentation classifier for imbalanced fault classification, X. Jiang and, Z. Ge, IEEE Transactions on Automation Science and Engineering, vol. 18, no. 3, pp. 1206–1217, , 2020

    21. Deactivation of cobalt based Fischer–Tropsch catalystsa review, Ø. Borg, E. Rytter and, N. E. Tsakoumis, A. Holmen, M. Rønning, vol. 154, no. 3-4, pp. 162–182, , 2010

    22. Hybrid modeling approach for terpolymerization reactions in cstr, S. H. Choi and, K. Lee, J. M. Lee, IFAC-PapersOnLine, vol. 55, no. 7, pp. 266– 271, , 2022

    23. Scale-up of microchannel reactors for fischer- tropsch synthesis, J. J. Lerou and, D. R. Kilanowski, K. T. Jarosch, S. R. Deshmukh, S. P. Fitzgerald, L. Schrader, T. J. Mazanec, A. L. Y. Tonkovich, vol. 49, no. 21, pp. 10883–10888, 2010., , 2010

    24. CFD-based multi-objective optimization of a quench reactor design, B. Meyer, A. Richter and, U. Prüfert, P. Rößger, K. Uebel, Fuel Processing Technology, vol. 149, pp. 290–304, , 2016

    25. On parameter estimation for semi-linear errors-in-variables models, C. Hengjian and, L. Rongcai, vol. 64, no. 1, pp. 1–24, , 1998

    26. Activation of precipitated iron Fischer-Tropsch synthesis catalysts, A. G. Sault and, N. B. Jackson, M. S. Harrington, K. E. Coulter, A. K. Datye, D. S. Kalakkad, S. D. Kohler, M. D. Shroff, vol. 156, no. 2, pp. 185–207, 1995., , 1995

    27. Co and terpolymer reactivity ratios of chemically amplified resists, N. S. Pujari, M. Wang and, K. E. Gonsalves, Polymer, vol. 118, pp. 201–214, , 2017

    28. Kinetics of Fischer– Tropsch synthesis on titania-supported cobalt, C. H. Bartholomew, M. Tagliabue and, R. Zennaro, vol. 58, no. 4, pp. 309–319, , 2000

    29. Resolution and etch resistance of a family of 193nm positive resists, R. DiPietro, R. Kunz, D. Hofer and, R. Allen, I. Wan, G. Wallraff, vol. 8, no. 4, pp. 623–626, , 1995

    30. Learning from class-imbalanced dataReview of methods and applications, G. Mingyun, G. Bing, J. Shang, G. Haixiang, H. Yuanyue and, L. Yijing, Expert Systems With Applications, vol. 73, pp. 220–239, , 2017

    31. Refractive index prediction models for polymers using machine learning, R. Ramprasad, C. Kim, L. Chen, R. Batra and, J. P. Lightstone, vol. 127, no. 21, , 2020

    32. Flow properties of abs (acrylonitrile–butadiene–styrene) terpolymer, H. Kubota, vol. 19, no. 8, pp. 2299– 2308, , 1975

    33. Correlation and prediction of the refractive indices of polymers by qspr, M. Karelson, A. R. Katritzky, S. Sild and, vol. 38, no. 6, pp. 1171–1176, , 1998

    34. Methods for plant data-based process modeling in soft-sensor development, D. Sliškovi´c, Ž. Hocenski, R. Grbi´c and, Automatika, vol. 52, no. 4, pp. 306–318, , 2011

    35. Robust and consistent estimation of nonlinear errors-in-variables models, T. Li, vol. 110, no. 1, pp. 1–26, , 2002

    36. On-line principal component analysis with application to process modeling, W. Yu, T. Chai and, L. Zhao, J. Tang, vol. 82, pp. 167– 178, , 2012

    37. Qspr studies on refractive indices of structurally heterogeneous polymers, A. A. Toropov, S. E. Fioressi, L. M. Saavedra, A. P. Toropova and, P. R. Duchowicz, D. E. Bacelo, Chemometrics and Intelligent Laboratory Systems, vol. 140, pp. 86–91, , 2015

    38. Terpolymerization of acrylonitrile styrene and 2 3-dibromopropyl acrylate, Z. Janovic and, O. Vogl, K. Saric, Journal of Polymer Science: Polymer Chemistry Edition, vol. 21, no. 7, pp. 1913–1928, , 1983

    39. Carbonate looping experiments in a 1 mwth pilot plant and model validation, B. Epple, J. Kremer, J. Ströhle, M. Junk, A. Galloy and, vol. 127, pp. 13–22, , 2014

    40. Kinetics of free-radical copolymerization of α- methylstyrene and styrene, A. Rudin and, S. Chiang, Journal of Polymer Science: Polymer Chemistry Edition, vol. 12, no. 10, pp. 2235–2254, , 1974

    41. A recursive nonlinear pls algorithm for adaptive nonlinear process modeling, G. Wang and, J. Zhang, H. Ye, C. Li, Chemical Engineering & Technology: Industrial Chemistry-Plant Equipment-Process Engineering- Biotechnology, vol. 28, no. 2, pp. 141–152, , 2005

    42. High refractive index polymersfundamental research and practical applications, J.-g. Liu and, M. Ueda, vol. 19, no. 47, pp. 8907–8919, , 2009

    43. Soft sensor for melt index prediction based on long short-term memory network, J. M. Lee, M. J. Song, S. Kim, S. H. Oh, P. S. Jo and, IFACPapersOnLine, vol. 55, no. 7, pp. 857–862, , 2022

    44. A novel dynamic pca algorithm for dynamic data modeling and process monitoring, S. J. Qin, Y. Dong and, vol. 67, pp. 1–11, , 2018

    45. A weighted auto regressive lstm based approach for chemical processes modeling, S. Xu, X. Zhang, S. Li and, Y. Zou, Neurocomputing, vol. 367, pp. 64–74, , 2019

    46. Demystifying the estimation of reactivity ratios for terpolymerization systems, T. A. Duever and, N. Kazemi, A. Penlidis, vol. 60, no. 5, pp. 1752–1766, , 2014

    47. Importance of optimal composition in random terpolymer-based polymer solar cells, W. Lee, T. E. Kang, B. J. Kim, H.-H. Cho, H. Kang and, H. J. Kim, Macromolecules, vol. 46, no. 17, pp. 6806–6813, , 2013

    48. Syngas production for gas-to-liquids applicationstechnologies issues and outlook, A. Karp and, R. Dickenson, D. Simbeck, D. Wilhelm, Fuel Processing Technology, vol. 71, no. 1-3, pp. 139–148, , 2001

    49. Refractive indices of diverse data set of polymersA computational qspr based study, P. Boudjouk, F. Jabeen, M. Chen, M. Ossowski and, B. Rasulev, vol. 137, pp. 215–224, , 2017

    50. Effect of styrene acrylonitrile copolymer on the performance of tep/pmma composites, X. Yuan, K. Zhang, J. Mei, D. Wang and, vol. 8, no. 7, p. 075102, , 2021

    51. Prediction of refractive indices of linear polymers by a four-descriptor qspr model, Q. Zhang and, B. Chen, J. Xu, B. Guo, Polymer, vol. 45, no. 26, pp. 8651–8659, , 2004

    52. Support vector machine based modeling of an industrial natural gas sweetening plant, H. Adib, N. Mehranbod, M. Koolivand, F. Sharifi, N. M. Kazerooni and, vol. 14, pp. 121–131, , 2013

    53. CFD simulation of a packed bed industrial absorber with interbed liquid distributors, C.-H. Lee and, M. Oh, D.-H. Oh, H.-T. Oh, D. Q. Gbadago, vol. 95, p. 102983, , 2020

    54. Regret for expected improvement over the best-observed value and stopping condition in, V. Nguyen, S. Rana, S. Gupta, C. Li and, S. Venkatesh, Asian Conference on Machine Learning, pp. 279–294, PMLR, , 2017

    55. Reconciling modern machinelearning practice and the classical bias–variance trade-off, D. Hsu, S. Mandal, S. Ma and, M. Belkin, vol. 116, no. 32, pp. 15849–15854, , 2019

    56. A general framework for constrained bayesian optimization using information-based search, R. P. Adams, M. W. Hoffman and, J. M. Hernández-Lobato, M. A. Gelbart, Z. Ghahramani, vol. 17, pp. 1–53, , 2016

    57. Laminated leak-resistant chemical processors methods of making and methods of operating., J. Marco, R. Luzenski, A. Tonkovich, J. Marco, M. Marchiando, T. Yuschak, S. Deshmukh and, P. Neagle, L. W. Keyes, 2019/0224639 A1, , 2019

    58. Miscibility of blends of polymers based on styrene acrylonitrile and methyl methacrylate, H. Keskkula and, D. Paul, M. Nishimoto, Polymer, vol. 30, no. 7, pp. 1279–1286, , 1989

    59. Selective olefin production on silica based iron catalysts in Fischer–Tropsch synthesis, H.-J. Chun, Y. T. Kim, J.-C. Seo, E.-h. Sim, S. Yang, S. Lee, J. Lee and, Catalysis Science & Technology, vol. 12, no. 19, pp. 5814–5828, , 2022

    60. A flexible transfer learning framework for bayesian optimization with convergence guarantee, S. Venkatesh, S. Gupta and, T. T. Joy, S. Rana, Expert Systems with Applications, vol. 115, pp. 656–672, , 2019

    61. A stopping criterion for bayesian optimization by the gap of expected minimum simple regrets, H. Ishibashi, I. Takeuchi and, H. Hino, M. Karasuyama, International Conference on Artificial Intelligence and Statistics, pp. 6463–6497, PMLR, 2023, , 2023

    62. Influence of molecular weight on the fracture properties of aliphatic polyketone terpolymers, R. Gaymans, H. Huétink and, D. Homminga, W. Zuiderduin, Polymer, vol. 44, no. 20, pp. 6361–6370, , 2003

    63. The maximum likelihood solution of the problem of estimating a linear functional relationship, M. E. Solari, Journal of the Royal Statistical Society: Series B (Statistical Methodology), vol. 31, no. 2, pp. 372–375, , 1969

    64. and Fischer–Tropsch catalysts for the production of hydrocarbon fuels with high selectivity, K. Cheng, J. Kang, Q. Zhang, Y. Wang, W. Deng, vol. 7, no. 5, pp. 1251–1264, , 2014

    65. Benchmarking multi-objective bayesian global optimization strategies for aerodynamic design in, K. Shimoyama, C. Amalinadhi, P. S. Palar and, Y. B. Dwianto, L. R. Zuhal, 2018 AIAA/ASCE/AHS/ASC Structures, Structural Dynamics, and Materials Conference, p. 0914, , 2018

    66. Impeller shapeoptimization of stirred-tank reactorCFD and fluid structure interaction analyses, G. Najafi, S. Hoseini, B. Ghobadian and, A. Akbarzadeh, vol. 413, p. 127497, , 2021

    67. Refractive index control in bicomponent polymer films for integrated thermooptical applications, F. Horowitz, M. R. Becker, R. A. F. Dau, M. B. Pereira and, P. A. Soave, Optical Engineering, vol. 48, no. 12, pp. 124603– 124603, , 2009

    68. Scale-up of microand milli-reactorsAn overview of strategies design principles and applications, F. Zhao, Z. Wen, T. Noël, Z. Dong, S. Kuhn and, Chemical Engineering Science: X, vol. 10, p. 100097, , 2021

    69. Binary vs. ternary reactivity ratiosAppropriate estimation procedures with terpolymerization data, A. J. Scott and, A. Penlidis, vol. 105, pp. 442–450, , 2018

    70. Multi-objective bayesian optimization of chemical reactor design using computational fluid dynamics, J. M. Lee, M. Kim and, S. Park, J. Na, Computers & Chemical Engineering, vol. 119, pp. 25–37, , 2018

    71. Decision tree and first-principles model-based approach for reactor runaway analysis and forecasting, F. Szeifert and, T. Varga, J. Abonyi, Engineering Applications of Artificial Intelligence, vol. 22, no. 4-5, pp. 569–578, , 2009

    72. First-principles data-based and hybrid modeling and optimization of an industrial hydrocracking unit, N. Bhutani, A. Ray, G. Rangaiah and, vol. 45, no. 23, pp. 7807–7816, , 2006

    73. Overfitting or perfect fitting? risk bounds for classification and regression rules that interpolate, D. J. Hsu and, P. Mitra, M. Belkin, Advances in Neural Information Processing Systems, vol. 31, , 2018

    74. Reactivity ratios for the terpolymerization of methyl methacrylate vinyl acetate and molecular oxygen, P. De and, D. Sathyanarayana, Journal of Polymer Science Part A: Polymer Chemistry, vol. 40, no. 4, pp. 564–572, , 2002

    75. Scalable meta-learning for bayesian optimization using ranking-weighted gaussian process ensembles in, E. Bakshy, M. Feurer, B. Letham and, AutoML Workshop at International Conference on Machine Learning, vol. 7, , 2018

    76. Batch-to-batch optimal control of a batch polymerisation process based on stacked neural network models, J. Zhang, vol. 63, no. 5, pp. 1273–1281, , 2008

    77. CFD simulation of crossflow mixing in a packed bed using porous media model and experimental validation, E. Specht, A. Alkhalaf and, K. Mohammadpour, vol. 6, pp. 157–162, , 2019

    78. Machine learning approach for prediction of the grafting yield in radiation-induced graft polymerization, Y. Maekawa, N. Seko and, Y. Ueki, vol. 25, p. 101158, , 2021

    79. Parameter estimation and prediction uncertainties for multi-response kinetic models with uncertain inputs, K. Abdi, K. B. McAuley, B. Celse and, vol. 69, no. 6, p. e18058, 2023, , 2023

    80. Multiscale modeling of fixed-bed reactors with porous (open-cell foam) non-spherical particlesHydrodynamics, N. Deen and, S. Das, J. Kuipers, vol. 334, pp. 741–759, , 2018

    81. Stopping criteria for boosting automatic experimental design using real-time fmri with bayesian optimization, G. Montana, A. A. Faisal, R. P. Monti, R. Leech and, C. Anagnostopoulos, R. Lorenz, I. R. Violante, arXiv preprint arXiv:1511.07827, , 2015

    82. The macroscopic equations for simultaneous heat and mass transfer in isotropic continuous and closed systems, H. Merk, Applied Scientific Research, Section A, vol. 8, pp. 73–99, , 1959

    83. Design of sustainable chemical processesSystematic retrofit analysis generation and evaluation of alternatives, H. Matos, A. Carvalho, R. Gani and, Process Safety and Environmental Protection, vol. 86, no. 5, pp. 328– 346, , 2008

    84. Data augmentation and machine learning techniques for control strategy development in bio-polymerization process, S. Wei, S. K. Arumugasamy and, Z. Chen, I. M. L. Chew, vol. 11, p. 100172, , 2022

    85. Reactivity ratio estimation in multicomponent polymerizations using the error-in-variables-model (EVM) framework, N. Kazemi, PhD thesis University of Waterloo, , 2014

    86. Application of latent variable methods to process control and multivariate statistical process control in industry, T. Kourti, International Journal of Adaptive Control and Signal Processing, vol. 19, no. 4, pp. 213–246, , 2005

    87. Synthesis of styrene-acrylonitrile copolymers and related block copolymers by atom transfer radical polymerization, T. Sarbu, N. V. Tsarevsky, K. Matyjaszewski, B. Göbelt and, Macromolecules, vol. 35, no. 16, pp. 6142– 6148, , 2002

    88. Microchannel reactor for Fischer–Tropsch synthesisAdaptation of a commercial unit for testing microchannel blocks, S. Yunes and, O. Sanz, J. D’olhaberriague, M. Montes, L. C. Almeida, vol. 110, pp. 171–177, , 2013

    89. Study of dielectric properties of styrene-acrylonitrile graphite sheets composites in low and high frequency region, B. Kang, S. Park and, J.-O. Park, V. Panwar, R. Mehra, vol. 45, no. 6, pp. 1777– 1784, , 2009

    90. Mathematical modeling of multicomponent chain-growth polymerizations in batch semibatch and continuous reactorsa review, A. Penlidis and, J. B. Soares, A. E. Hamielec, M. A. Dubé, vol. 36, no. 4, pp. 966–1015, , 1997

    91. The free-radical terpolymerisation of ethylene methyl acrylate and vinyl acetate at high pressure. polymerization tests, M. Dorn, F. Stein and, G. Luft, Die Angewandte Makromolekulare Chemie: Applied Macromolecular Chemistry and Physics, vol. 211, no. 1, pp. 131–140, , 1993

    92. Design of a microstructured reactor with integrated heat-exchanger for optimum performance of a highly exothermic reaction, E. Rebrov, J. Schouten, M. De Croon and, vol. 69, no. 1-4, pp. 183–192, , 2001

    93. Computational fluid dynamics study of heat transfer in a microchannel reactor for low-temperature Fischer–Tropsch synthesis, G. Arzamendi, P. M. Diéguez, L. M. Gandía, M. Montes, J. A. Odriozola, E. F. Sousa- Aguiar and, vol. 160, no. 3, pp. 915–922, , 2010

    94. Optimization of fischer–tropsch microchannel reactor using computational fluid dynamics and enveloped bayesian optimization, K. Lee and, J. M. Lee, p. 108658, 2024, , 2024

    95. Application of CFD on the optimization by response surface methodology of a micromixing unit and its use as a chemical microreactor, J. Ortega-Casanova, Chemical Engineering and Processing: Process Intensification, vol. 117, pp. 18–26, , 2017

    96. Exploring advanced process equipment visualization as a step towards digital twins development in the chemical industryA CFD-dnn approach, D. Q. Gbadago, S. Hwang, J. Moon and, vol. 40, no. 1, pp. 37–45, 2023, , 2023

    97. CFD modeling of a multichannel Fischer-Tropsch reactor module with microscale cooling channelsEffects of mirrored structure cooling layers, D. B. Oh, Y.-J. Lee and, Y. Woo, M.-J. Park, S. J. Han, J. E. Park, pp. 1–9, 2023, , 2023

    98. Multi-objective optimization of microchannel reactor for Fischer-Tropsch synthesis using computational fluid dynamics and genetic algorithm, C. Han, J. Na, U. Lee and, K. S. Kshetrimayum, vol. 313, pp. 1521–1534, , 2017

    99. A deep learning approach using temporal-spatial data of computational fluid dynamics for fast property prediction of gas- solid fluidized bed, Z. Xia and, L. Guo, P. Qin, Korean Journal of Chemical Engineering, vol. 40, no. 1, pp. 57–66, 2023, , 2023

    100. CFD simulation of microchannel reactor block for Fischer– Tropsch synthesisEffect of coolant type and wall boiling condition on reactor temperature, K. S. Kshetrimayum, C. Han, I. Jung, C.-J. Lee and, S. Park, Y. Lee, S. Park, J. Na, vol. 55, no. 3, pp. 543–554, , 2016

    101. Explicit residence time distribution of a generalised cascade of continuous stirred tank reactors for a description of short recirculation time (bypassing), P. Doshi and, D. Jajcevic, P. Toson, Processes, vol. 7, no. 9, p. 615, , 2019

    102. Online quality prediction of nonlinear and non-gaussian chemical processes with shifting dynamics using finite mixture model based gaussian process regression approach, J. Yu, vol. 82, pp. 22–30, , 2012

    더보기

    분석정보

    View

    상세정보조회

    0

    Usage

    원문다운로드

    0

    대출신청

    0

    복사신청

    0

    EDDS신청

    0

    동일 주제 내 활용도 TOP

    더보기

    주제

    연도별 연구동향

    연도별 활용동향

    연관논문

    연구자 네트워크맵

    공동연구자 (7)

    유사연구자 (20) 활용도상위20명

    이 자료와 함께 이용한 RISS 자료

    나만을 위한 추천자료

    해외이동버튼