«Automation and Informatization of the fuel and energy complex»

Identification of partial failures in gas transportation system using a neural network algorithm

UDC: 004.942:621.644.8
DOI: -

Authors:

¹ Gazprom Promgaz, St. Petersburg, Russia
² National University of Oil and Gas "Gubkin University", Moscow, Russia

Keywords: neural network algorithms, gas transportation system, partial failures, mode modeling, hydraulic efficiency coefficient

Annotation:

An artificial neural network has been built to recognize critical situations in the operating gas transportation system (GTS). To form the training sample, data on the GTS actual operational modes were used as well as the results of hydraulic calculations obtained by modeling off-nominal situations on a programming and computing suite. Before carrying out the corresponding calculations, the model was adjusted taking into account the operational characteristics of the system for the selected lead period. To characterize the state of pipelines, "defining points" of the system were found, which made it possible to several times reduce the dimensionality of input data array of the neural network without losing the quality of its work. Partial failures of the GTS were simulated by reducing the hydraulic efficiency factors in several sections of the gas pipelines during the modeling process. As a result, a multilayer perceptron model was built, which, receiving the data on pressure and flow rate measurements at the "defining points" of the GTS as input ones, provides information at the output with the help of which it is possible to determine the presence and location of a critical situation on pipelines. Options for further development of the proposed neural network algorithm are outlined. In particular, to take into account changes of the system topology over time, it is proposed to additionally use information on the position of the block valve stations.

Bibliography:

1. Cheng D. Verification, Validation, Uncertainty Quantification Issues in the Development of Artificial Intelligent Models in Flow Assurance // PSIG Annual Meeting, Virtual, May 3–7, 2021. – Paper No. PSIG-2113.
2. O novom metode tsifrovogo modelirovaniya nestatsionarnykh rezhimov techeniya gaza v magistral'nykh gazoprovodakh s primeneniem neyronnykh operatorov / A.V. Belinskiy, D.V. Gorlov, I.A. Pyatyshev, A.E. Titov // Gazovaya prom-st'. – 2024. – № 5(865). – S. 54–66.
3. Prakticheskaya realizatsiya primeneniya neyrosetevykh modeley dlya modelirovaniya slozhnykh gazotransportnykh sistem bol'shoy razmernosti / N.A. Kislenko, S.N. Pankratov, V.V. Dedkova [i dr.] // Avtomatizatsiya i informatizatsiya TEK. – 2024. – № 9(614). – S. 33–41.
4. Sukharev M.G., Tverskoy I.V. O metodologii primeneniya matematicheskikh modeley neyronnykh setey k problemam neftegazovogo kompleksa // Avtomatizatsiya i informatizatsiya TEK. – 2022. – № 2(583). – S. 28–35. – DOI: 10.33285/2782-604X-2022-2(583)-28-35
5. Zhang Chi, Shafieezadeh A. Nested physics-informed neural network for analysis of transient flows in natural gas pipelines // Engineering Applications of Artificial Intelligence. – 2023. – Vol. 122. – P. 106073. – DOI: 10.1016/j.engappai.2023.106073
6. Salmasi F., Khatibi R., Ghorbani M.A. A study of friction factor formulation in pipes using artificial intelligence techniques and explicit equations // Turkish J. of Engineering and Environmental Sciences. – 2012. – Vol. 36, No. 2. – P. 121–138. – DOI: 10.3906/muh-1008-30
7. Learning of viscosity functions in rarefied gas flows with physics-informed neural networks / J.-M. Tucny, M. Durve, A. Montessori, S. Succi // Computers & Fluids. – 2024. – Vol. 269. – P. 106114. – DOI: 10.1016/j.compfluid.2023.106114
8. Chel L., Busi L., Carcasci C. Offline Monitoring Method for a Natural Gas City Gate Station Odorization System // PSIG Annual Meeting, San Diego, California, USA, May 10–13, 2022. – Paper No. PSIG-2213.
9. A systematic hybrid method for real-time prediction of system conditions in natural gas pipeline networks / Su Huai, E. Zio, Zhang Jinjun [et al.] // J. of Natural Gas Science & Engineering. – 2018. – Vol. 57. – P. 31–44. – DOI: 10.1016/j.jngse.2018.06.033
10. Brown R., Resto A. Proactive Parametric Studies Drive System Intelligence and Practical Gas Business Benefits // PSIG Annual Meeting, San Diego, California, USA, May 10–13, 2022. – Paper No. PSIG-2203.
11. Sardanashvili S.A. Raschetnye metody i algoritmy (truboprovodnyy transport gaza). – M.: Izd-vo "Neft' i gaz" RGU nefti i gaza im. I.M. Gubkina, 2005. – 577 s.
12. STO Gazprom 2-3.5-051-2006. Normy tekhnologicheskogo proektirovaniya magistral'nykh gazoprovodov. – Vved. 2006–07–03. – M.: IRTs Gazprom, 2006. – VIII, 197 s.
13. Sukharev M.G., Samoylov R.V. Analiz i upravlenie statsionarnymi i nestatsionarnymi rezhimami transporta gaza: monogr. – M.: Izdat. tsentr RGU nefti i gaza (NIU) im. I.M. Gubkina, 2016. – 399 s.
14. STO Gazprom 2-3.5-433-2010. Metodika po provedeniyu gidravlicheskikh raschetov i opredeleniyu tekhnicheski vozmozhnoy proizvoditel'nosti ekspluatiruemykh sistem magistral'nykh gazoprovodov. – Vved. 2010–08–09. – M.: Gazprom, 2010. – IV, 28 s.
15. Levenberg K. A Method for the Solution of Certain Non-Linear Problems in Least Squares // Quarterly of Applied Mathematics. – 1944. – Vol. 2. – P. 164–168. – DOI: 10.1090/qam/10666
16. Marquardt D. An Algorithm for Least-Squares Estimation of Nonlinear Parameters // J. of the Society for Industrial and Applied Mathematics. – 1963. – Vol. 11, Issue 2. – P. 431–441. – DOI: 10.1137/0111030