Sign in Subscribe
Machine learning

Robust Physical Indicator for Leak Detection in Urban Water Pipelines

Strategic Insights

— 1 min read

A new study by researchers from Tongji University proposes a robust physical indicator for identifying leaks in urban water pipelines, grounded in the physical background of leakage noise sources. The research aims to improve the accuracy of leakage detection methods, which currently rely heavily on data-driven features. The study conducted experiments to validate the effectiveness and robustness of the proposed indicator, achieving recognition accuracies of 99.89% for Support Vector Machine (SVM) and 99.97% for eXtreme Gradient Boosting (XGBoost).

Key Takeaways:

  • The study proposes a novel physics-informed indicator for leak detection in urban water pipelines, which addresses the limitation of existing methods that overly rely on data-driven features.
  • Experiments were conducted to validate the effectiveness and robustness of the proposed indicator, achieving recognition accuracies of 99.89% for SVM and 99.97% for XGBoost.
  • The research was funded by the National Key R&D Program of China and the National Natural Science Foundation of China.
  • The proposed indicator is grounded in the physical background of leakage noise sources, and a rigorous theoretical analysis leads to the development of an effective physical indicator.
  • The field test conducted on an in-service water supply pipeline with a total length of 701 m achieved recognition accuracies of 97.92% for SVM and 99.31% for XGBoost.
  • The research was published in the journal Sensors, with a free version available at https://doi-org.sdpl.idm.oclc.org/10.3390/s25165069.

Statistics:

  • Recognition accuracies of 99.89% for SVM and 99.97% for XGBoost were achieved in the experiments.
  • The field test achieved recognition accuracies of 97.92% for SVM and 99.31% for XGBoost, respectively.
  • The study was funded by the National Key R&D Program of China and the National Natural Science Foundation of China.

Sources:

  • Novel Physics-Informed Indicators for Leak Detection in Water Supply Pipelines. Sensors, 2025,25(16):5069. (Sensors - http://www.mdpi.com/journal/sensors).
  • The publisher for Sensors is MDPI AG.
  • NewsRx. Study Findings from Tongji University Advance Knowledge in Sensor Research (Novel Physics-Informed Indicators for Leak Detection in Water Supply Pipelines). Robotics & Machine Learning. September 8, 2025; p 587.

Read more