Dynamic Prediction of Natural Gas Calorific Value Based on Deep Learning

Jingjing Hu, Zhaoming Yang () and Huai Su ()
Additional contact information
Jingjing Hu: National Engineering Research Center of Oil and Gas Pipeline Transportation Safety, Beijing Key Laboratory of Urban Oil and Gas Distribution Technology, China University of Petroleum−Beijing, Beijing 102249, China
Zhaoming Yang: National Engineering Research Center of Oil and Gas Pipeline Transportation Safety, Beijing Key Laboratory of Urban Oil and Gas Distribution Technology, China University of Petroleum−Beijing, Beijing 102249, China
Huai Su: National Engineering Research Center of Oil and Gas Pipeline Transportation Safety, Beijing Key Laboratory of Urban Oil and Gas Distribution Technology, China University of Petroleum−Beijing, Beijing 102249, China

Energies, 2023, vol. 16, issue 2, 1-20

Abstract: The natural gas quality fluctuates in complex natural gas pipeline networks, because of the influence of the pipeline transmission process, changes in the gas source, and fluctuations in customer demand in the mixing process. Based on the dynamic characteristics of the system with large time lag and non−linearity, this article establishes a deep−learning−based dynamic prediction model for calorific value in natural gas pipeline networks, which is used to accurately and efficiently analyze the dynamic changes of calorific value in pipeline networks caused by non−stationary processes. Numerical experiment results show that the deep−learning model can effectively extract the effects of non−stationary and large time lag hydraulic characteristics on natural gas calorific value distribution. The method is able to rapidly predict the dynamic changes of gas calorific value in the pipeline network, based on real−time operational data such as pressure, flow rate, and gas quality parameters. It has a prediction accuracy of over 99% and a calculation time of only 1% of that of the physical simulation model (built and solved based on TGNET commercial software). Moreover, with noise and missing key parameters in the data samples, the method can still maintain an accuracy rate of over 97%, which can provide a new method for the dynamic assignment of calorific values to complex natural gas pipeline networks and on−site metering management.

Keywords: natural gas pipeline network; deep learning; calorific value prediction; dynamic prediction (search for similar items in EconPapers)
JEL-codes: Q Q0 Q4 Q40 Q41 Q42 Q43 Q47 Q48 Q49 (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
https://www.mdpi.com/1996-1073/16/2/799/pdf (application/pdf)
https://www.mdpi.com/1996-1073/16/2/799/ (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gam:jeners:v:16:y:2023:i:2:p:799-:d:1031052

Access Statistics for this article

Energies is currently edited by Ms. Agatha Cao

More articles in Energies from MDPI
Bibliographic data for series maintained by MDPI Indexing Manager ().