Comparative Evaluation of Prognostic Models for Medium-Term Pressure Prediction in Gas Transmission Units
##plugins.themes.bootstrap3.article.main##
##plugins.themes.bootstrap3.article.sidebar##
https://orcid.org/0000-0001-8184-8238
Kamal MEDJAHER
Abstract
Gas pressure reduction and delivery stations are critical components of gas transmission networks, where pressure drifts may signal early degradation or control malfunction. Despite the availability of measurements supporting data-driven Prognostics and Health Management (PHM), medium-term forecasting remains difficult due to non-stationarities, operator adjustments, and heterogeneous setpoints. This study investigates 5-day-ahead pressure forecasting for maintenance planning and early detection of evolutions preceding threshold exceedances. A robust preprocessing pipeline addresses irregular sampling, missing data, and level shifts under real operating conditions. Two interpretable models are evaluated: SARIMAX, for seasonal and exogenous effects, and LightGBM, for nonlinear dynamics with feature-importance analysis. Performance is assessed on a homogeneous subset of gas pressure measurement recorders using standard regression metrics. The results emphasize not only predictive capability but also interpretability, positioning transparent forecasting models as scalable and auditable building blocks for PHM in gas transmission networks.
How to Cite
##plugins.themes.bootstrap3.article.details##
Prognostics and Health Management, time series forecasting, predictive maintenance, gas pressure monitoring, non-stationary time series
Chalgham, W., Wu, K.-Y., & Mosleh, A. (2020, December). System-level prognosis and health monitoring modeling framework and software implementation for gas pipeline system integrity management. Journal of Natural Gas Science and Engineering, 84, 103671.
Ding, J., Zhao, Y., & Jin, J. (2023, May). Forecasting natural gas consumption with multiple seasonal patterns. Applied Energy, 337, 120911.
Jie, H., Yuan, C., Liu, H., Liu, Q., Qiu, X., Zhang, M., & Zhao, M. (2025, April). Deep learning-based signal processing and prediction for gas pressure regulating stations. Journal of Physics: Conference Series, 012055.
Lim, B., & Zohren, S. (2021, April). Time series forecasting with deep learning: A survey. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, 379(2194).
Lu, H., & Frank Cheng, Y. (2025, September). Artificial intelligence in energy pipelines: Opportunities and risks. Engineering.
Magdin, M. (2025, December). Literature review: Current trends and advances in the use of artificial intelligence for ensuring the safety and efficiency of gas pipeline operations. Results in Engineering, 28, 107309.
Tian, N., Shao, B., Bian, G., Zeng, H., Li, X., & Zhao, W. (2024, March). Application of forecasting strategies and techniques to natural gas consumption: A comprehensive review and comparative study. Engineering Applications of Artificial Intelligence, 129, 107644.
Wang, B., Guo, Y., Wang, D., Zhang, Y., He, R., & Chen, J. (2022, December). Prediction model of natural gas pipeline crack evolution based on optimized DCNN-LSTM. Mechanical Systems and Signal Processing, 181, 109557.
Zhao, W., Shao, B., Tian, N., Zhang, W., Zhao, X., & Wu, S. (2025, February). Early warning study of field station process safety based on VMD-CNN-LSTM-self-attention for natural gas load prediction. Scientific Reports, 15(1), 6360.
This work is licensed under a Creative Commons Attribution 3.0 Unported License.
The Prognostic and Health Management Society advocates open-access to scientific data and uses a Creative Commons license for publishing and distributing any papers. A Creative Commons license does not relinquish the author’s copyright; rather it allows them to share some of their rights with any member of the public under certain conditions whilst enjoying full legal protection. By submitting an article to the International Conference of the Prognostics and Health Management Society, the authors agree to be bound by the associated terms and conditions including the following:
As the author, you retain the copyright to your Work. By submitting your Work, you are granting anybody the right to copy, distribute and transmit your Work and to adapt your Work with proper attribution under the terms of the Creative Commons Attribution 3.0 United States license. You assign rights to the Prognostics and Health Management Society to publish and disseminate your Work through electronic and print media if it is accepted for publication. A license note citing the Creative Commons Attribution 3.0 United States License as shown below needs to be placed in the footnote on the first page of the article.
First Author et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 3.0 United States License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.