Intelligent prediction of flammable gas dispersion from urban gas pipeline leakage using physics-informed neural networks

Runquan Li, Xinhong Li and Ahmed Salim

Reliability Engineering and System Safety, 2026, vol. 268, issue C

Abstract: This study develops an PINNs based intelligent model for predicting gas dispersion from urban gas pipelines leakage. Time-series data of gas concentrations during pipeline leakage event are collected using a sparsely distributed sensor array surrounding pipeline and are used as inputs to a fully connected neural network. By leveraging AD, the model computes the spatial and temporal gradients of gas concentration and incorporates physical governing equations including the continuity equation, momentum equation, and component transport equation into loss function. A combined optimization objective function is formulated by integrating data regression loss and physical residual loss. Through loss minimization, the trained model not only achieves accurate data fitting but also strictly adheres to the physical laws governing fluid dispersion. Experimental results indicate that the model (R2 = 0.9968, MSE = 4.38 × 10–6, RMSE = 0.0021, MAE = 0.0017) outperforms traditional methods in both prediction accuracy and generalization capability. The developed model can support early warning and emergency response in urban gas pipeline leakage.

Keywords: Urban gas pipeline leakage; Gas dispersion; Concentrations; Intelligent prediction; PINNs (search for similar items in EconPapers)
Date: 2026
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0951832025012372
Full text for ScienceDirect subscribers only

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:eee:reensy:v:268:y:2026:i:c:s0951832025012372

DOI: 10.1016/j.ress.2025.112038

Access Statistics for this article

Reliability Engineering and System Safety is currently edited by Carlos Guedes Soares

More articles in Reliability Engineering and System Safety from Elsevier
Bibliographic data for series maintained by Catherine Liu ().