Advanced numerical techniques for characterising obstructions in sewer pipes

The underground sewer system in the U.K. is approximately 300,000 km long, for which the replacement costs are estimated to be 104 billion. The sewer system is owned by the privatised water companies who have a legal duty to maintain the structural and operational conditions of their sewer systems, and this includes reducing flooding incidents. In approximately 80% of cases, flooding incidents are caused by obstructions arising from the deterioration of a pipe wall, or from large deposits of sediment and/or fat. The detection and removal of obstructions should form part of any maintenance programme, although the ability to do this is currently restricted by the lack of a fast and reliable method. This project will focus on using sound waves to detect and characterise obstructions in sewers. Here, loudspeakers generate a pressure pulse that travels down a sewer pipe; this pulse is normally strongly reflected by any obstruction it encounters and by using microphones to capture the reflected energy information about the obstruction may be captured quickly and easily. Accordingly, this method offers a fast and objective way to monitor large sewer systems.The proposed research aims to deliver a step change improvement to a prototype acoustic device developed in a previous (experimentally based) EPSRC project (EP/D058589/1). The current device relies on cross-correlation between new acoustic intensity measurements and measurements stored for known sewer defects; however, this methodology is limited by the number of experimental studies it's possible to undertake and difficulties when interpreting measured intensity data. Furthermore, the current method can say nothing about the geometry, or surface characteristics, of an obstruction, and there is no proof that a unique link exists between the measured data and the properties of the obstruction. The proposed research seeks to address these issues by using mathematical models to aid in the development of a new measurement methodology that treats the acoustic intensity as a complex quantity rather than using the traditional real valued representation adopted in the current device. Here, complex acoustic intensity has the potential to uncover significantly more information from scattered sound fields when compared to a real valued intensity representation, and it is the measurement of complex intensity in acoustic waveguides that forms the focus of this proposal.Although complex intensity measurements have the potential to deliver significantly more information, they are not well understood, especially for scattering from obstacles in an acoustic waveguides. Accordingly, to gain a better understanding of complex intensity it is desirable to develop mathematical models and here both frequency and time domain models are proposed. The frequency domain model is based on the finite element method in order to accommodate those irregular geometries typically found in sewer systems; the time domain model is based on taking an inverse Fourier transform of the frequency domain calculations and will also utilise an inverse analysis in order to address issues such as the uniqueness of measured data. Theoretical predictions will be compared with time-averaged and instantaneous complex intensity measurements obtained under laboratory conditions. In this way, a more general understanding of complex intensity will be developed before this knowledge is applied to the development of a new measurement methodology for sewer systems. Furthermore, to maintain relevance to real sewer systems problems known to affect the accuracy of field measurements, such as manholes, cracks, joints and pipe surface roughness will also be studied. Accordingly, the understanding developed with the mathematical models and laboratory measurements will be used to develop a new prototype experimental methodology suitable for reconstructing the geometry and surface characteristics of obstructions in real sewer systems.