ایجاد حسگر نرم تطبیقی برای پایش به‌هنگام ترکیب درصد گاز: پیاده‌سازی در برج تقطیر آزئوتروپیک فرایند تولید مونومر وینیل‌استات

نویسندگان

گروه طراحی فرایند، دانشکده مهندسی شیمی، دانشگاه تربیت مدرس، تهران، ایران

چکیده
موضوع تحقیق: تحلیل به‌هنگام به کمک ابزارهای دیجیتال، نیازمند دریافت در لحظه داده‌ها از نقاط مختلف فرایندهای شیمیایی صنعتی است. تأخیرهای زمانی در اندازه‌گیری متغیرهای فرایندی ممکن است عملکرد مؤثر راهبرد‌های کنترلی مختلف، پایداری فرایند و کارایی عملیاتی را تحت تأثیر قرار دهد و در نتیجه تحلیل، استخراج اطلاعات و تبدیل آن به تصمیم‌های قابل اجرا در لحظه را غیرممکن ‌کند. فرایند تولید مونومر وینیل‌استات (VAM) به‌عنوان فرایند پیچیده و غیرخطی در صنایع شیمیایی شناخته می‌شود. در این فرایند ترکیب درصد آب در پایین برج تقطیر آزئوتروپیک، یکی از متغیرهای مهم است که توسط دستگاه کروماتوگرافی گازی (GC) با تأخیر زمانی و هزینه زیاد اندازه‌گیری می‌شود.

روش تحقیق: حسگرهای نرم به‌طور عمده تخمین به‌هنگام متغیرهایی که اندازه‌گیری آن‌ها دشوار یا غیرممکن است را بهبود می‌بخشند. شبکه‌های عصبی به واسطه قابلیت یادگیری الگوهای غیرخطی و سرعت پیش‌بینی مناسب در توسعه حسگرهای نرم نقش مهمی را ایفا می‌کنند. این پژوهش بر روی توسعه حسگر نرم تطبیقی بر اساس مدل شبکه عصبی پیش‌خور برای تخمین به‌هنگام ترکیب درصد آب در پایین برج تقطیر آزئوتروپیک در فرایند VAM متمرکز است.

نتایج اصلی: حسگر نرم ایجاد شده به‌صورت تطبیقی در حضور عیب‌های عملیاتی مختلف به کار گرفته شده است و با خطای MSE=1.1x10-5، رفتار GC را به صورت لحظه‌ای تخمین می‌زند. می‌توان با حفظ دقت پیش‌بینی در پیاده‌سازی تطبیقی حسگر نرم در حضور تغییرات مختلف فرایندی، سازگاری مؤثر این حسگرها را اثبات کرد. این پژوهش قابلیت حسگرهای نرم را به‌عنوان جایگزینی کارآمد و مقرون‌به‌صرفه برای نظارت به‌هنگام در فرایندهای شیمیایی پیچیده را نشان می‌دهد.

کلیدواژه‌ها

موضوعات


عنوان مقاله English

Development of an adaptive soft sensor for real-time monitoring of gas composition: Implementation on the azeotropic distillation column of the vinyl acetate monomer production process

نویسندگان English

Amir Arsalan Sobhani
Mohammad fakhroleslam
Process Design Department, Faculty of Chemical Engineering, Tarbiat Modares University, Tehran, Iran
چکیده English

Research subject: Real-time analysis using digital tools requires receiving instantaneous data from various points in industrial chemical processes. Time delays in measurements of process variables can affect the effective performance of different control strategies, process stability, and operational efficiency, making it impossible to analyze, extract information, and convert it into actionable decisions in real-time. The synthesis process of vinyl acetate monomer is recognized as a benchmark dynamic and nonlinear process in the chemical industry. In this process, the composition of water at the bottom of the azeotropic distillation column is one of the important variables measured by a gas chromatography (GC) analyzer, which has a significant time delay and high cost.

Research approach: Soft sensors primarily improve the real-time estimation of variables that are difficult or impossible to measure. Neural networks play an important role in the development of soft sensors due to their ability to learn nonlinear patterns and their suitable prediction speed. This study focuses on the development of a soft sensor based on a feedforward neural network model for real-time estimation of the composition of water at the bottom of an azeotropic distillation column in the vinyl acetate monomer synthesis process.

Main results: Additionally, the model was adaptively implemented under various fault conditions and accurately estimates the GC analyzer behavior instantaneously, achieving a mean squared error (MSE) of 1.1 × 10-5. Maintaining prediction accuracy in the adaptive implementation of soft sensors in the presence of various process faults demonstrates the effective adaptability of these sensors. Therefore, this study demonstrates the capability of soft sensors as an efficient and cost-effective alternative for real-time monitoring of complex chemical processes.

کلیدواژه‌ها English

Soft Sensor
GC analyzer
Neural Network
Azeotropic distillation
VAM process
[1] P. Fantke et al., “Transition to sustainable chemistry through digitalization,” Chem, vol. 7, no. 11, pp. 2866–2882, Nov. 2021, doi: 10.1016/j.chempr.2021.09.012.
[2] P. Mohindru, “A REVIEW ON SMART SENSORS USED IN CHEMICAL INDUSTRY 4.0”.
[3] E. H. C. Ito, A. R. Secchi, M. V. C. Gomes, and C. R. Paiva, “Development of a gas composition soft sensor for distillation columns: A simplified model based and robust approach,” in Computer Aided Chemical Engineering, vol. 44, Elsevier, 2018, pp. 661–666. doi: 10.1016/B978-0-444-64241-7.50105-1.
[4] S. P. H. R. Rajesh, “The Analysis of Different Types of IoT Sensors and security trend as Quantum chip for Smart City Management,” IOSR J. Bus. Manag., vol. 20, no. 1, pp. 55–60, Jan. 2018, doi: http://dx.doi.org/10.9790/487X-2001045560.
[5] Dr. Diane Turner, “Gas Chromatography – How a Gas Chromatography Machine Works, How To Read a Chromatograph and GCxGC,” Technology networks, Analysis and Separations. [Online]. Available: https://www.anthias.co.uk/
[6] T. Okada, H. Kaneko, and K. Funatsu, “Development of a model selection method based on the reliability of a soft sensor model,” 2012.
[7] Y. Jiang, S. Yin, J. Dong, and O. Kaynak, “A Review on Soft Sensors for Monitoring, Control, and Optimization of Industrial Processes,” IEEE Sens. J., vol. 21, no. 11, pp. 12868–12881, Jun. 2021, doi: 10.1109/JSEN.2020.3033153.
[8] P. Kadlec, B. Gabrys, and S. Strandt, “Data-driven Soft Sensors in the process industry,” Comput. Chem. Eng., vol. 33, no. 4, pp. 795–814, Apr. 2009, doi: 10.1016/j.compchemeng.2008.12.012.
[9] C. Abeykoon, “Design and Applications of Soft Sensors in Polymer Processing: A Review,” IEEE Sens. J., vol. 19, no. 8, pp. 2801–2813, Apr. 2019, doi: 10.1109/JSEN.2018.2885609.
[10] X. Yuan, S. Qi, Y. A. W. Shardt, Y. Wang, C. Yang, and W. Gui, “Soft sensor model for dynamic processes based on multichannel convolutional neural network,” Chemom. Intell. Lab. Syst., vol. 203, p. 104050, Aug. 2020, doi: 10.1016/j.chemolab.2020.104050.
[11] I. A. Udugama et al., “Novel Soft Sensor for Measuring and Controlling Product Recovery in a High-Purity, Multicomponent, Side-Draw Distillation Column,” Ind. Eng. Chem. Res., vol. 58, no. 43, pp. 20026–20035, Oct. 2019, doi: 10.1021/acs.iecr.9b04594.
[12] S. Park and C. Han, “A nonlinear soft sensor based on multivariate smoothing procedure for quality estimation in distillation columns,” Comput. Chem. Eng., vol. 24, no. 2–7, pp. 871–877, Jul. 2000, doi: 10.1016/S0098-1354(00)00343-4.
[13] R. F. E. Jiménez, C. M. A. Zaragoza, J. A. Hernández, M. A. Medina, and G. V. G. Ramírez, “DISEÑO E IMPLEMENTACIÓN DE UN SENSOR VIRTUAL BASADO EN OBSERVADOR PARA UN INTERCAMBIADOR DE,” 2011.
[14] W. Zhu, Y. Ma, Y. Zhou, M. Benton, and J. Romagnoli, “Deep Learning Based Soft Sensor and Its Application on a Pyrolysis Reactor for Compositions Predictions of Gas Phase Components,” in Computer Aided Chemical Engineering, vol. 44, Elsevier, 2018, pp. 2245–2250. doi: 10.1016/B978-0-444-64241-7.50369-4.
[15] A. Rani, V. Singh, and J. R. P. Gupta, “Development of soft sensor for neural network based control of distillation column,” ISA Trans., vol. 52, no. 3, pp. 438–449, May 2013, doi: 10.1016/j.isatra.2012.12.009.
[16] K. Funatsu, “Process Control and Soft Sensors,” in Applied Chemoinformatics, 1st ed., T. Engel and J. Gasteiger, Eds., Wiley, 2018, pp. 571–584. doi: 10.1002/9783527806539.ch13.
[17] P. Kadlec, R. Grbić, and B. Gabrys, “Review of adaptation mechanisms for data-driven soft sensors,” Comput. Chem. Eng., vol. 35, no. 1, pp. 1–24, Jan. 2011, doi: 10.1016/j.compchemeng.2010.07.034.
[18] L. P. A. Espufia, “A Systematic Approach for Soft Sensor Development”.
[19] B. Lin, B. Recke, J. K. H. Knudsen, and S. B. Jørgensen, “A systematic approach for soft sensor development,” Comput. Chem. Eng., vol. 31, no. 5–6, pp. 419–425, May 2007, doi: 10.1016/j.compchemeng.2006.05.030.
[20] M. T. Tham, G. A. Montague, A. Julian Morris, and P. A. Lant, “Soft-sensors for process estimation and inferential control,” J. Process Control, vol. 1, no. 1, pp. 3–14, Jan. 1991, doi: 10.1016/0959-1524(91)87002-F.
[21] B. M. Wise, N. B. Gallagher, S. Watts Butler, D. D. White, and G. G. Barna, “Development and Benchmarking of Multivariate Statistical Process Control Tools for a Semiconductor ETCH Process: Impact of Measurement Selection and Data Treatment on Sensitivity,” IFAC Proc. Vol., vol. 30, no. 18, pp. 35–42, Aug. 1997, doi: 10.1016/S1474-6670(17)42377-9.
[22] R. Chen, K. Dave, T. J. McAvoy, and M. Luyben, “A Nonlinear Dynamic Model of a Vinyl Acetate Process,” Ind. Eng. Chem. Res., vol. 42, no. 20, pp. 4478–4487, Oct. 2003, doi: 10.1021/ie020859k.
[23] Y. Dodge, The concise encyclopedia of statistics. in Springer reference. New York: Springer, 2010.
[24] W.-L. Loh, “ON LATIN HYPERCUBE SAMPLING”.
[25] X. Yuan, Y. Gu, Y. Wang, C. Yang, and W. Gui, “A Deep Supervised Learning Framework for Data-Driven Soft Sensor Modeling of Industrial Processes,” IEEE Trans. Neural Netw. Learn. Syst., vol. 31, no. 11, pp. 4737–4746, Nov. 2020, doi: 10.1109/TNNLS.2019.2957366.
[26] Y.-L. He, Z.-Q. Geng, and Q.-X. Zhu, “Data driven soft sensor development for complex chemical processes using extreme learning machine,” Chem. Eng. Res. Des., vol. 102, pp. 1–11, Oct. 2015, doi: 10.1016/j.cherd.2015.06.009.
[27] H. Pan, T. Su, X. Huang, and Z. Wang, “LSTM-based soft sensor design for oxygen content of flue gas in coal-fired power plant,” Trans. Inst. Meas. Control, vol. 43, no. 1, pp. 78–87, Jan. 2021, doi: 10.1177/0142331220932390.
[28] Y. S. Perera, D. A. A. C. Ratnaweera, C. H. Dasanayaka, and C. Abeykoon, “The role of artificial intelligence-driven soft sensors in advanced sustainable process industries: A critical review,” Eng. Appl. Artif. Intell., vol. 121, p. 105988, May 2023, doi: 10.1016/j.engappai.2023.105988.
[29] D. Rodriguez-Granrose et al., “Design of experiment (DOE) applied to artificial neural network architecture enables rapid bioprocess improvement,” Bioprocess Biosyst. Eng., vol. 44, no. 6, pp. 1301–1308, Jun. 2021, doi: 10.1007/s00449-021-02529-3.