Non-Destructive Characterization of Cured-in-Place Pipe Defects

Area of Science:

• Civil Engineering
• Materials Science
• Geophysics

Background:

• Urban sewage and water networks are critical infrastructure requiring regular maintenance.
• Trenchless rehabilitation, especially cured-in-place pipe (CIPP) technology, is vital for urban areas.
• Current diagnostic methods like camera scans and laser scans lack material characterization capabilities.

Purpose of the Study:

• To introduce and evaluate three innovative non-destructive testing (NDT) methods for characterizing pipe conditions.
• To assess the effectiveness of impact-echo, ground-penetrating radar (GPR), and impedance spectroscopy for detecting defects in polymer-lined pipes.
• To compare traditional and deep learning machine learning algorithms for defect characterization using impact-echo data.

Main Methods:

• Impact-echo method combined with deep learning on continuous wavelet transform images for defect characterization.
• Ground-penetrating radar (GPR) with a heuristic algorithm for detecting caverns behind pipes.
• Impedance spectroscopy for characterizing polymer liner delamination due to uneven curing.

Main Results:

• The study successfully characterized delamination, identified caverns behind CIPP, and evaluated overall pipe health using the novel NDT methods.
• A deep learning algorithm demonstrated effectiveness in defect characterization from impact-echo signals.
• GPR and impedance spectroscopy provided accurate estimations of subsurface defects and liner delamination, respectively.

Conclusions:

• The presented NDT methods offer enhanced accuracy and material characterization capabilities for assessing urban pipe networks compared to traditional techniques.
• These innovative methods are crucial for effective maintenance and rehabilitation planning of essential water and sewage infrastructure.
• The integration of machine learning with NDT data provides a powerful tool for automated defect analysis and condition assessment.