Model-Free Matched Field Processing for Condition Assessment of Pressurized Pipes

Many transient-based defect detection methods are based on matching the transient model to the measured data to localize the defects. Despite the high degree of success in laboratory systems, this approach has not been successful in the field because the system information needed to carry out proper modeling is often lacking as records are often unavailable or incomplete. In this paper, a method using matched-field processing (MFP) is applied to data versus data (DD) to localize defects in pressurized freshwater pipe systems. The proposed MFP-DD method does not require transient models and relies only on the measured reference transient response of the system, which is then correlated with the more recent transient response to localize the defects. The method requires (1) repeatable transient tests to obtain the model-free MFP results and (2) topological information of the system to interpret these results. The proposed method is demonstrated numerically and validated through experiments in laboratory, field-scale transmission-like pipeline, and real-field distribution systems in Hong Kong where a simulated leak is successfully localized. In addition, the method proved its strength by identifying a modification in the real-field distribution system through a blind test. This modification was made by the water authority officers and was unbeknown to the authors.