Passive imaging of water pipelines using ambient turbulence noise

Ambient noise generated by flowing water turbulence is harnessed as a signal source for imaging key parameters and fault detection in water pipelines. This approach is important because it can aid in the estimation of wave speed or detection of water pipeline defects such as blockages and leakages. More importantly it overcomes the challenging problem of generating a signal source of sufficient power to provide the necessary signal-to-noise ratios for conventional water pipeline imaging and fault detection techniques. In this paper, the expressions of the auto- and cross-correlation functions of the ambient noise between sensors are derived using wave theory. It is shown that the time-domain Green's functions can be extracted from the correlation functions. Experimental and numerical examples are provided for water pipelines to demonstrate that wave speed can be estimated from the time-domain Green's functions. A method for extending the technique, by using straightforward but accurate approximations of the correlation functions, to detect the presence of defects in the profile of the water pipeline is also proposed.