A Digital Twin-Driven Multisource Fault Diagnosis System for Smart Water Distribution Networks

Jinjiang Zhao; Yinfeng Wang; Lei Zhang

doi:10.54097/sf5f2p82

Authors

Jinjiang Zhao Institute of Automation, Gansu Academy of Sciences, Lanzhou Gansu 730000, China；College of Electrical and Information Engineering, Lanzhou University of Technology, Lanzhou Gansu 730050, China
Yinfeng Wang Shaanxi Provincial Water and Drought Disaster Prevention Center, Xi’an Shaanxi 710004, China
Lei Zhang Institute of Automation, Gansu Academy of Sciences, Lanzhou Gansu 730000, China； College of Electrical and Information Engineering, Lanzhou University of Technology, Lanzhou Gansu 730050, China

DOI:

https://doi.org/10.54097/sf5f2p82

Keywords:

Robust control; sensor bias; multisensor fusion; digital twin; fault diagnosis; engineering stability.

Abstract

Due to the combination of sensing, hydraulics, computation and operation in the digital twin fault diagnosis of smart water distribution networks, there are many sources of uncertainty, and thus a single deterministic controller or classifier cannot be used in engineering practice. The main contents of this paper are a strong diagnosis and control system, multi-sensor fusion, bias-aware residual construction, model-predictive reasoning, conservative fallback rules, etc. Parametric scenario data will be employed in the framework evaluation, and all evaluation indices and threshold parameters will be set to guarantee the repeatability of the engineering test. The five tables present the sensor channel, disturbance case, diagnostic threshold, control response and robustness indicator; the two analytical formulas are the residual score and the weighted stability index. Based on the above tests, it can be seen that the proposed framework has a small false alarm rate and is relatively stable under all circumstances; even when the input signal is biased, delayed or partially missing, the chain of evidence connecting fault data, model constraint derivation and control measures can still meet the audit requirements of standard engineering journals.

Downloads

Download data is not yet available.

References

[1] Uribe, Z. M., Galván, R. R., Alvarado, Á. M. J., et al. (2026). Artificial intelligence in water distribution networks: A systematic review of models, input variables, databases, and output strategies for leak detection. Smart Cities, 9(3), 45. https://doi.org/10.3390/smartcities9030045.

[2] Bam, G. P., Rezaei, N., Roubanis, A., et al. (2025). Digital twin applications in the water sector: A review. Water, 17(20), 2957. https://doi.org/10.3390/w17202957

[3] Vanja, T., Emanuel, G., Ivana, L., et al. (2025). Digital twin for a real-time leakage detection and localization in pressurized piping systems. Journal of Hydroinformatics, 27(4), 755–770. https://doi.org/10.2166/hydro.2025.304

[4] Chatterjee, S., Ghumkar, S., Ahbab, M. M., et al. (2026). Early detection of DMA-level leaks in water networks using robust regression ensemble framework. Water, 18(5), 563. https://doi.org/10.3390/w18050563

[5] Cognata, L. R., Piazza, S., & Freni, G. (2025). Bridging the gap between model assumptions and realities in leak localization for water networks. Water, 17(24), 3502. https://doi.org/10.3390/w17243502

[6] Peng, S., Zeng, H., Wu, X., et al. (2025). Leakage break diagnosis for water distribution network using LSTM-FCN neural network based on high-frequency pressure data. Water, 17(12), 1823. https://doi.org/10.3390/w17121823

[7] Menapace, A., Zanfei, A., Herrera, M., et al. (2024). Graph neural networks for sensor placement: A proof of concept towards a digital twin of water distribution systems. Water, 16(13), 1835. https://doi.org/10.3390/w16131835

[8] Fu, A., Qiao, R., & Wu, Z. (2024). Decentralized periodic event-triggered control for smart water distribution systems via digital twin approach. International Journal of Robust and Nonlinear Control, 34(14), 9638–9653. https://doi.org/10.1002/rnc.7484

[9] Mucke, N. T., Pandey, P., Jain, S., et al. (2023). A probabilistic digital twin for leak localization in water distribution networks using generative deep learning. Sensors, 23(13), 6179. https://doi.org/10.3390/s23136179

[10] Guangtao, F., Yiwen, J., Siao, S., et al. (2022). The role of deep learning in urban water management: A critical review. Water Research, 223, 118973. https://doi.org/10.1016/j.watres.2022.118973.