In-line Nondestructive Inspection of Mechanical Defects in Pipelines with Shear Horizonital Wave EMAT

Intelligent Automation, Inc. and Oak Ridge National Lab is proposing a novel and integrated approach to inspect mechanical damages in the pipelines with or without coatings. It extends our successful work in a previous phase 1 project that combines the state of the art Shear Horizontal wave signal processing and pattern classification techniques, to detect and characterize the mechanical dents in the underground pipeline transportation infrastructures. In particular, this proposal address several challenging issues: 1) thorough guided wave modal analysis through three dimensional Boundary Element Method for best wave mode and frequency selection and defect volumetric feature extraction, 2) circumferential and axial ultrasonic SH wave EMATs and fixture design for data collection and analysis, 3) advanced signal processing algorithms such as Principal Component Analysis and Discriminant Analysi for mechanical defect signature extraction and pattern classification, and 4) small dents and coating condition assessment. The envisioned system will be mounted on a PIG for accurately and reliably detecting the metal loss and other defects in the pipeline infrastructures long ahead before they need remediation.