An Acoustic Method for Condition Classification in Live Sewer Networks

Underground pipes are an important part of urban water infrastructure. These pipes are gradually deteriorating due to aging, operational stresses and environmental conditions. In order to be able to manage the underground pipe system efficiently, condition monitoring is needed to provide a clear understanding of the behavior of sewer systems under various hydraulic conditions. This paper reports on the application of a novel acoustic method to study the evolution of blockages and various types of damage in a full scale life sewer pipe which has been installed in the hydraulic laboratory at the University of Bradford. Temporal and frequency characteristics in the behavior of the acoustic intensity are extracted from the acoustic signals recorded on an array of microphones. These characteristics are used for pattern recognition which is based on K-nearest neighbors (KNN) classifier. The obtained results indicate that the pattern recognition system can provide a reliable classification of the pipe condition in the presence and absence of flow.