Газовая Промышленность 3.2023

Научная статья

УДК 004.942::665.6
(UDK 004.942::665.6)

Для получения доступа к статьям

Авторизуйтесь

АВТОМАТИЗАЦИЯ (AUTOMATION (INCLUDING ROBOTICS FOR OIL AND GAS INDUSTRY, SOFTWARE SYSTEMS, DATA ANALYSIS, ETC.))

ОПЫТ ПРИМЕНЕНИЯ ИМИТАЦИОННОГО МОДЕЛИРОВАНИЯ ДЛЯ ПОВЫШЕНИЯ ТОЧНОСТИ ОЦЕНКИ КАЧЕСТВА КОНЕЧНОГО ПРОДУКТА

(EXPERIENCE OF USING SIMULATIONS TO IMPROVE EVALUATION ACCURACY OF FINAL PRODUCT QUALITY)

В настоящее время топливно-энергетические ресурсы составляют основную статью расходов на предприятиях нефтегазопереработки и нефтехимии России. В связи с этим одна из важных задач сегодня – повышение эффективности таких производств с минимальными энергетическими затратами.
При управлении непрерывными технологическими процессами широкое распространение получили виртуальные анализаторы. Они не только используются как самостоятельная надстройка для оценки качества выходных продуктов, но и интегрируются в систему усовершенствованного управления технологическими процессами для повышения их энергоэффективности, что коррелирует с основными положениями концепции Индустрии 4.0. Главная проблема разработки виртуальных анализаторов заключается в наличии шумов, например вызванных неточными наблюдениями или неисправностью датчиков, отсутствии значений параметров различных режимов функционирования промышленных установок, а также нестационарности исследуемого процесса.
В статье показана возможность использования при построении виртуальных анализаторов совмещенных данных (имитационного моделирования и промышленных) в условиях их плохой репрезентативности, связанной с отсутствием информации во всем диапазоне функционирования промышленной установки. Такой подход позволяет повысить точность виртуальных анализаторов. В работе представлены результаты их построения на совмещенных данных для оценки показателя качества выходного продукта массообменного технологического процесса в целях решения проблемы плохой репрезентативности обучающей выборки. В качестве примера рассматривается процесс синтеза высокооктановой добавки бензинов – метил-трет-бутилового эфира.

Fuel and energy resources are currently the main expense items at oil and gas refineries and petrochemical plants in Russia. Therefore, it is extremely important now to improve the efficiency of such plants while minimizing energy costs.
Soft sensors are widely used in control of continuous industrial processes. They act as customizing add-in for evaluating the finished product quality, and also integrate into advanced process control systems to improve their energy efficiency, which correlates with the main provisions of the Industry 4.0 concept. The main problems in the development of soft sensors are: noise (e. g. caused by inaccurate observations or faulty sensors), lack of various operation mode values for industrial units, and nonstationary nature of the process under investigation.
The article demonstrates the possibility to use the combined data (simulation and industrial) when setting-up soft sensors at poor data representativeness due to lack of information within the entire unit operation range. This approach enables improving the accuracy of soft sensors. The article presents soft sensors’ setup results based on the combined data to evaluate the quality of the final product of a mass-exchange process aimed at solving the training set poor representativeness issue. Synthesis of high-octane additive to gasoline (methyl tert-butyl ether) was analyzed as an example.

ВИРТУАЛЬНЫЙ АНАЛИЗАТОР, ИМИТАЦИОННОЕ МОДЕЛИРОВАНИЕ, УПРАВЛЕНИЕ, ТЕХНОЛОГИЧЕСКИЙ ОБЪЕКТ, НЕЙРОННАЯ СЕТЬ, МЕТИЛ-ТРЕТ-БУТИЛОВЫЙ ЭФИР, РЕПРЕЗЕНТАТИВНОСТЬ ОБУЧАЮЩЕЙ ВЫБОРКИ, РАСШИРЕНИЕ ВЫБОРКИ

SOFT SENSOR, SIMULATION, CONTROL, PROCESS FACILITY, NEURAL NETWORK, METHYL TERT-BUTYL ETHER, TRAINING SET REPRESENTATIVENESS, SET EXPANSION

С.А. Шевлягина, к.т.н., лауреат Международного конкурса молодых ученых «Нефтегазовые проекты: взгляд в будущее», ФГБУН Институт автоматики и процессов управления Дальневосточного отделения Российской академии наук (Владивосток, Россия), samotylova@dvo.ru

S.A. Shevlyagina, PhD in Engineering, winner of the International Young Scientists Awards “Oil and Gas Projects: A Glance into the Future”, Institute of Automation and Control Processes, Far Eastern Branch of Russian Academy of Sciences (Vladivostok, Russia), samotylova@dvo.ru

Vairo T., Reverberi A.P., Bragatto P.A., et al. Predictive model and soft sensors application to dynamic process operative control // Chemical Engineering Transactions. 2021. Vol. 86. P. 535–540. DOI: 10.3303/CET2186090.

Jiang Y., Yin S., Dong J., Kaynak O. A review on soft sensors for monitoring, control, and optimization of industrial processes // IEEE Sensors Journal. 2021. Vol. 21, No. 11. P. 12868–12881. DOI: 10.1109/JSEN.2020.3033153.

Мусаев А.А. Виртуальные анализаторы: концепция построения и применения в задачах управления непрерывными технологическими процессами // Автоматизация в промышленности. 2003. № 8. С. 28–33.

Бахтадзе Н.Н. Виртуальные анализаторы (идентификационный подход) // Автоматика и телемеханика. 2004. № 11. С. 3–24.

Shokry A., Vicente P., Escudero G., et al. Data-driven soft-sensors for online monitoring of batch processes with different initial conditions // Comput. Chem. Eng. 2018. Vol. 118. P. 159–179. DOI: 10.1016/j.compchemeng.2018.07.014.

Zhu J., Ge Z., Song Z., Gao F. Review and big data perspectives on robust data mining approaches for industrial process modeling with outliers and missing data // Annual Reviews in Control. 2018. Vol. 46. P. 107–133. DOI: 10.1016/j.arcontrol.2018.09.003.

Andrijić Ž.U., Cvetnić M., Bolf N. Soft sensor models for a fractionation reformate plant using small and bootstrapped data sets // Braz. J. Chem. Eng. 2018. Vol. 35, No. 2. P. 745–756. DOI: 10.1590/0104-6632.20180352s20150727.

Zhu Q.X., Zhang X.H., He Y.L. Novel virtual sample generation based on locally linear embedding for optimizing the small sample problem: Case of soft sensor applications // Ind. Eng. Chem. Res. 2020. Vol. 59, No. 40. P. 17977–17986. DOI: 10.1021/acs.iecr.0c01942.

Ditzler G., Roveri M., Alippi C., Polikar R. Learning in nonstationary environments: A survey // IEEE Computational Intelligence Magazine. 2015. Vol. 10, No. 4. P. 12–25. DOI: 10.1109/MCI.2015.2471196.

Ko C., Lee H., Lim Y., Lee W.B. Development of augmented virtual reality-based operator training system for accident prevention in a refinery // Korean Journal of Chemical Engineering. 2021. Vol. 38, No. 8. P. 1566–1577. DOI: 10.1007/s11814-021-0804-6.

Hsiao Y.D., Kang J.L., Wong D.S.H. Development of robust and physically interpretable soft sensor for industrial distillation column using transfer learning with small datasets // Processes. 2021. Vol. 9, No. 4. Article ID 667. DOI: 10.3390/pr9040667.

Sharma N., Liu Y.A. A hybrid science-guided machine learning approach for modeling chemical processes: A review // AIChE J. 2022. Vol. 68, No. 5. Article ID e17609. DOI: 10.1002/aic.17609.

Ahmad I., Ayub A., Kano M., Cheema I.I. Gray-box soft sensors in process industry: Current practice, and future prospects in era of big data // Processes. 2020. Vol. 8, No. 2. Article ID 243. DOI: 10.3390/pr8020243.

Nentwich C., Winz J., Engell S. Surrogate modeling of fugacity coefficients using adaptive sampling // Ind. Eng. Chem. Res. 2019. Vol. 58, No. 40. P. 18703–18716. DOI: 10.1021/acs.iecr.9b02758.

Nentwich C., Engell S. Surrogate modeling of phase equilibrium calculations using adaptive sampling // Comput. Chem. Eng. 2019. Vol. 126. P. 204–217. DOI: 10.1016/j.compchemeng.2019.04.006.

Bouaswaig A.E., Rahimi-Adli K., Roth M., et al. Application of a grey-box modelling approach for the online monitoring of batch production in the chemical industry // at-Automatisierungstechnik. 2020. Vol. 68, No. 7. P. 582–598. DOI: 10.1515/auto-2020-0038.

Самотылова С.А., Климченко В.В., Торгашов А.Ю. Применение адаптивных моделей в системах усовершенствованного управления массообменными технологическими объектами // Вестник Дальневосточного отделения Российской академии наук. 2021. № 4 (218). С. 48–52. DOI: 10.37102/0869-7698_2021_218_04_04.

Luo N., Qian F., Ye Z.C., et al. Estimation of mass-transfer efficiency for industrial distillation columns // Ind. Eng. Chem. Res. 2012. Vol. 51, No. 7. P. 3023–3031. DOI: 10.1021/ie2008407.

Mashunin Yu.K. Solving composition and decomposition problems of synthesis of complex engineering systems by vector-optimization methods // Journal of Computer and Systems Sciences International. 1999. Vol. 38, No. 3. P. 421–426.

Samotylova S.A., Torgashov A.Yu. Developing a soft sensor for MTBE process based on a small sample // Autom. Remote Control (Engl. Transl.). 2020. Vol. 81, No. 11. P. 2132–2142. DOI: 10.1134/S0005117920110120.

Самотылова С.А., Торгашов А.Ю. Алгоритм построения виртуального анализатора показателя качества выходного продукта ректификационной колонны в условиях малой обучающей выборки // Известия Санкт-Петербургского государственного технологического института (технического университета). 2019. № 48 (74). С. 36–41.
 
Sundmacher K., Uhde G., Hoffmann U. Multiple reactions in catalytic distillation processes for the production of the fuel oxygenates MTBE and TAME: Analysis by rigorous model and experimental validation // Chem. Eng. Sci. 1999. Vol. 54, No. 13–14. P. 2839–2847. DOI: 10.1016/S0009-2509(98)00520-X.

Samotylova S.A., Torgashov A.Yu. Application of a first principles mathematical model of a mass-transfer technological process to improve the accuracy of the estimation of the end product quality // Theor. Found. Chem. Eng. 2022. Vol. 56, No. 3. P. 371–387. DOI: 10.1134/S0040579522020117.

Torgashov A., Samotylova S., Yang F. Soft sensors development for industrial reactive distillation processes under small training datasets // Comput.-Aided Chem. Eng. 2022. Vol. 49. P. 937–942. DOI: 10.1016/B978-0-323-85159-6.50156-1.

Vairo T, Reverberi AP, Bragatto PA, Milazzo MF, Fabiano B. Predictive model and soft sensors application to dynamic process operative control. Chemical Engineering Transactions. 2021; 86: 535–540. https://doi.org/10.3303/CET2186090.

Jiang Y, Yin S, Dong J, Kaynak O. A review on soft sensors for monitoring, control, and optimization of industrial processes. IEEE Sensors Journal. 2021; 21(11): 12868–12881. https://doi.org/10.1109/JSEN.2020.3033153.

Musaev AA. Soft sensors: The concept of development and application in the continuous industrial processes control. Automation in Industry [Avtomatizatsiya v promyshlennosti]. 2003; (8): 28–33. (In Russian)

Bakhtadze NN. Virtual analyzers: Identification approach. Automation and Remote Control [Avtomatika i telemehanika]. 2004; (11): 3–24. (In Russian)

Shokry A, Vicente P, Escudero G, Pérez-Moya M, Graells M, Espuña A. Data-driven soft-sensors for online monitoring of batch processes with different initial conditions. Comput. Chem. Eng. 2018; 118: 159–179. https://doi.org/10.1016/j.compchemeng.2018.07.014.

Zhu J, Ge Z, Song Z, Gao F. Review and big data perspectives on robust data mining approaches for industrial process modeling with outliers and missing data. Annual Reviews in Control. 2018; 46: 107–133. https://doi.org/10.1016/j.arcontrol.2018.09.003.

Andrijić ŽU, Cvetnić M, Bolf N. Soft sensor models for a fractionation reformate plant using small and bootstrapped data sets. Braz. J. Chem. Eng. 2018; 35(2): 745–756. https://doi.org/10.1590/0104-6632.20180352s20150727.

Zhu QX, Zhang XH, He YL. Novel virtual sample generation based on locally linear embedding for optimizing the small sample problem: Case of soft sensor applications. Ind. Eng. Chem. Res. 2020; 59(40): 17977–17986. https://doi.org/10.1021/acs.iecr.0c01942.

Ditzler G, Roveri M, Alippi C, Polikar R. Learning in nonstationary environments: A survey. IEEE Computational Intelligence Magazine. 2015; 10(4): 12–25. https://doi.org/10.1109/MCI.2015.2471196.

Ko C, Lee H, Lim Y, Lee WB. Development of augmented virtual reality-based operator training system for accident prevention in a refinery. Korean Journal of Chemical Engineering. 2021; 38(8): 1566–1577. https://doi.org/10.1007/s11814-021-0804-6.

Hsiao YD, Kang JL, Wong DSH. Development of robust and physically interpretable soft sensor for industrial distillation column using transfer learning with small datasets. Processes. 2021; 9(4): article ID 667. https://doi.org/10.3390/pr9040667.

Sharma N, Liu YA. A hybrid science-guided machine learning approach for modeling chemical processes: A review. AIChE J. 2022; 68(5): article ID e17609. https://doi.org/10.1002/aic.17609.

Ahmad I, Ayub A, Kano M, Cheema II. Gray-box soft sensors in process industry: Current practice, and future prospects in era of big data. Processes. 2020; 8(2): article ID 243. https://doi.org/10.3390/pr8020243.

Nentwich C, Winz J, Engell S. Surrogate modeling of fugacity coefficients using adaptive sampling. Ind. Eng. Chem. Res. 2019; 58(40): 18703–18716. https://doi.org/10.1021/acs.iecr.9b02758.

Nentwich C, Engell S. Surrogate modeling of phase equilibrium calculations using adaptive sampling. Comput. Chem. Eng. 2019; 126: 204–217. https://doi.org/10.1016/j.compchemeng.2019.04.006.

Bouaswaig AE, Rahimi-Adli K, Roth M, Hosseini A, Vale H, Engell S, et al. Application of a grey-box modelling approach for the online monitoring of batch production in the chemical industry. at-Automatisierungstechnik. 2020; 68(7): 582–598. https://doi.org/10.1515/auto-2020-0038.

Samotylova SA, Klimchenko VV, Torgashov AYu. Application of adaptive models in advanced mass transfer process control systems. Vestnik of Far Eastern Branch of Russian Academy of Sciences [Vestnik Dal’nevostochnogo otdeleniya Rossijskoj akademii nauk]. 2021; 218(4): 48–52. https://doi.org/10.37102/0869-7698_2021_218_04_04. (In Russian)

Luo N, Qian F, Ye ZC, Cheng H, Zhong WM. Estimation of mass-transfer efficiency for industrial distillation columns. Ind. Eng. Chem. Res. 2012; 51(7): 3023–3031. https://doi.org/10.1021/ie2008407.

Mashunin YuK. Solving composition and decomposition problems of synthesis of complex engineering systems by vector-optimization methods. Journal of Computer and Systems Sciences International. 1999; 38(3): 421–426.

Samotylova SA, Torgashov AYu. Developing a soft sensor for MTBE process based on a small sample. Autom. Remote Control (Engl. Transl.). 2020; 81(11): 2132–2142. https://doi.org/10.1134/S0005117920110120.

Samotylova SA, Torgashov AYu. Algorithm for soft sensor design of characteristic value of output product quality of distillation column in conditions a small training sample. Bulletin of the Saint Petersburg State Institute of Technology (Technical University) [Izvestiya Sankt-Peterburgskogo gosudarstvennogo tehnologicheskogo instituta (tehnicheskogo universiteta)]. 2019; 74(48): 36–41. (In Russian)

Sundmacher K, Uhde G, Hoffmann U. Multiple reactions in catalytic distillation processes for the production of the fuel oxygenates MTBE and TAME: Analysis by rigorous model and experimental validation. Chem. Eng. Sci. 1999; 54(13–14): 2839–2847. https://doi.org/10.1016/S0009-2509(98)00520-X.

Samotylova SA, Torgashov AYu. Application of a first principles mathematical model of a mass-transfer technological process to improve the accuracy of the estimation of the end product quality. Theor. Found. Chem. Eng. 2022; 56(3): 371–387. https://doi.org/10.1134/S0040579522020117.

Torgashov A, Samotylova S, Yang F. Soft sensors development for industrial reactive distillation processes under small training datasets. Comput.- Aided Chem. Eng. 2022; 49: 937–942. https://doi.org/10.1016/B978-0-323-85159-6.50156-1.

NEFTEGAS.info

Внимание к деталям — от идеи
до воплощения! Только актуальная информация и свежие новости.

Контакты

108811, г. Москва, Киевское ш.,
Бизнес-парк «Румянцево», корп. Б,
подъезд 5, офис 506 Б

+7 (495) 240-54-57