Leakage detection in a real distribution network using a SOM

Abstract

One great challenge for waterworks is effective leakage detection. This paper presents a method based on the self-organising map for leakage detection in a water distribution network. The data used for training and validating the test results consist of vectors of the flow meter readings and knowledge of reported leak locations. The most important factor facilitating the self-organising-map-based modelling of leaks is the developed leak function. The results of the experiments presented show that the model trained on flow data can detect leaks in a defined distribution network area.

Keywords:

• artificial neural networks
• flow
• leakage detection
• self-organising map
• water distribution network
• water loss

Acknowledgements

The Laboratory of Water and Wastewater Engineering of the Technical University of Helsinki wishes to acknowledge the support given for this research by the Finnish Funding Agency for Technology and Innovation (Tekes), Helsinki Water and all other participants in the project ‘The Development of Water and Energy Network Management’.