«Automation and Informatization of the fuel and energy complex»

On the possibility of using machine learning technology to simulate a compressor station of a main gas pipeline

UDC: 681.5:622.279
DOI: -

Authors:

¹ Gazprom transgaz Tchaikovsky, Tchaikovsky, Russia
² NIIgazekonomika, Moscow, Russia

Keywords: compressor station, mathematical modeling, dynamic system, machine learning, neural networks

Annotation:

The authors of the article examine the issues of developing a neural network model of a complex technical system – a compressor station, assessing the adequacy and feasibility of using such models when constructing digital twins of gas transportation systems fragments. It is assumed that such a model has a number of advantages, including allowing one to naturally take into account the data of the actual modes of the technological system and providing a quick way to solve a number of optimization problems. To form a neural network model, it is envisaged to use machine learning approaches on real or model data of a compressor station operation. The authors of the article propose several different neural network models and draw a conclusion about the best model for the technological system under study.

Bibliography:

1. Varlamov G.B., Priymak E.A. Fakticheskie kharakteristiki oborudovaniya kompressornykh stantsiy i optimizatsiya zagruzki magistral'nogo gazoprovoda // Vostochno-Evrop. zhurn. peredovykh tekhnologiy. – 2013. – T. 5, № 8(65). – S. 9–13.
2. Bulygina L.V. Optimizatsiya rezhimov raboty kompressornoy stantsii c tsel'yu povysheniya lokal'noy i sistemnoy energoeffektivnosti gazotransportnoy sistemy // Nauka i tekhnika v gazovoy prom-sti. – 2019. – № 1(77). – S. 53–68.
3. Kryukov O.V. Kompleksnaya optimizatsiya energopotrebleniya agregatov kompressornykh stantsiy // Ekspozitsiya Neft' Gaz. – 2015. – № 1(40). – S. 30–33.
4. Dynamic simulation of natural gas pipeline network based on interpretable machine learning model / Dengji Zhou, Xingyun Jia, Shixi Ma [et al] // Energy. – 2022. – Vol. 253. – P. 124068. – DOI: 10.1016/j.energy.2022.124068
5. Rodrigues L.R. A hybrid multi-population metaheuristic applied to load-sharing optimization of gas compressor stations // Computers & Electrical Engineering. – 2022. – Vol. 97. – P. 107632. – DOI: 10.1016/j.compeleceng.2021.107632
6. Energoeffektivnye rezhimy gazotransportnykh sistem i printsipy ikh obespecheniya / A.M. Karasevich, M.G. Sukharev, A.V. Belinskiy [i dr.] // Gazovaya prom-st'. – 2012. – № 1(672). – S. 30–34.
7. Vaszi Z., Szabó C., Varga A. Implementing the mathematical model of the throughput of compressor station aggregates // Nonlinear Analysis: Modelling and Control. – 2015. – Vol. 20, No. 2. – P. 291–304. – DOI: 10.15388/NA.2015.2.10
8. Botros K.K. Transient Phenomena in Compressor Stations During Surge // J. of Engineering for Gas Turbines and Power. – 1994. – Vol. 116, No. 1. – P. 133–142. – DOI: 10.1115/1.2906782
9. Pyanylo Ya.D., Gladun S.V. Optimization of energy costs for gas transportation in complex gas transmission systems // Annals of the Faculty of Engineering Hunedoara – Int. J. of Engineering. – 2015. – Vol. 13, No. 3. – P. 31–36.
10. Sukharev M.G., Kulalaeva M.A. Identification of model flow parameters and model coefficients with the help of integrated measurements of pipeline system operation parameters // Energy. – 2021. – Vol. 232. – P. 120864. – DOI: 10.1016/j.energy.2021.120864
11. Statisticheskaya matematicheskaya model' rascheta koeffitsienta poleznogo deystviya kompressorov turbodetandernykh agregatov: usovershenstvovanie i identifikatsiya / Yu.B. Galerkin, A.F. Rekstin, O.A. Solov'eva [i dr.] // Izv. vysshikh ucheb. zavedeniy. Mashinostroenie. – 2022. – № 7(748). – S. 68–81. – DOI: 10.18698/0536-1044-2022-7-68-81
12. Carter R.G. Compressor Station Optimization: Computational Accuracy and Speed // PSIG Annual Meeting, San Francisco, California, Oct. 23–25, 1996. – PSIG, 1996. – S. PSIG-9605.
13. Kazak A.S., Oleynikov A.V. Primenenie metodov sistemnogo analiza dlya postroeniya tsifrovykh dvoynikov fragmentov gazotransportnykh sistem na osnove neyrosetevogo modelirovaniya (osnovnye polozheniya metodologii analiza i sinteza) // Avtomatizatsiya i informatizatsiya TEK. – 2024. – № 1(606). – S. 5–12.
14. Issledovanie tselesoobraznosti neyrosetevogo modelirovaniya rezhimov raboty kompressornykh stantsiy / A.V. Oleynikov, A.S. Kazak, A.V. Belinskiy [i dr.] // Avtomatizatsiya i informatizatsiya TEK. – 2024. – № 2(607). – S. 10–19.
15. Zhenhao Tang, Zijun Zhang. The multi-objective optimization of combustion system operations based on deep data-driven models // Energy. – 2019. – Vol. 182. – P. 37–47. – DOI: 10.1016/j.energy.2019.06.051
16. Batayev N., Onbayev A. Prediction of Gas Turbine Parameters based on Machine Learning Regression Methods // The 6th Int. Virtual Conf. on Advanced Scientific Results. – 2018. – P. 217–221. – DOI: 10.18638/scieconf.2018.6.1.495
17. Applying Artificial Neural Network to Optimize the Performance of the Compressor Station: A Case Study / I. Đuračić, M. Stojkov, T. Šarić [et al.] // Tehnički vjesnik. – 2021. – Vol. 28, No. 4. – P. 1197–1202. – DOI: 10.17559/TV-20200926113750
18. Online Feedback Optimization of Compressor Stations with Model Adaptation using Gaussian Process Regression / M. Zagorowska, M. Degner, L. Ortmann [et al.] // J. of Process Control. – 2023. – Vol. 121. – P. 119–133. – DOI: 10.1016/j.jprocont.2022.12.001
19. Vilalta P.C., Hui Wan, Patnaik S.S. Centrifugal Compressor Performance Prediction Using Gaussian Process Regression and Artificial Neural Networks // ASME Int. Mechanical Engineering Congress and Exposition, Salt Lake City, Utah, USA, Nov. 11–14, 2019. – American Society of Mechanical Engineers, 2019. – DOI: 10.1115/IMECE2019-11936
20. Application of Gaussian Processes to online approximation of compressor maps for load-sharing in a compressor station / A. Ahmed, M. Zagorowska, E.A. del Rio-Chanona, M. Mercangöz // 2022 European Control Conf. (ECC), London, UK, July 12–15, 2022. – IEEE, 2022. – P. 205–212. – DOI: 10.23919/ECC55457.2022.9838042
21. Multiobjective Tree-structured Parzen Estimator for Computationally Expensive Optimization Problems / Yo. Ozaki, Yu. Tanigaki, S. Watanabe, M. Onishi // Proceedings of the 2020 genetic and evolutionary computation conf., Cancún, Mexico, July 8–12, 2020. – New York, NY, United States: Association for Computing Machinery, 2020. – P. 533–541. – DOI: 10.1145/3377930.3389817