Guided wave synthetic focusing imaging in limited access pipes for improved quantification accuracy

Abstract

Routine inspection of pipelines and processing and production plants is paramount for their safe and reliable operation. Detection and criticality assessment of defects appearing in inaccessible locations in pipelines poses a great challenge for many industries. Therefore, inspection methods capable of defect detection, location and accurate characterisation in these unreachable locations from a distance away are needed. Guided wave testing is a promising non-destructive testing technique to address inspection of such inaccessible areas. The method has been extensively deployed for rapid screening of large lengths of pipes from a single device position. Because the technique is designed for large volumetric coverage it provides very limited individual feature characterisation. The areas where potential defects are indicated are typically inspected locally using more accurate methods (e.g. phased array ultrasonic testing) to confirm defect presence and perform its characterisation. The follow-up inspections of the inaccessible locations are not possible, hence defect characterisation using guided wave testing needs to be improved. The research presented in this thesis aims to improve defect detection, location and characterisation in guided wave testing by the deployment of high resolution synthetic focusing imaging. The common source method (CSM), total focusing method (TFM) and plane wave imaging (PWI) algorithms are adapted to guided wave testing of pipes. Their performance is evaluated and compared in terms of lateral resolution and signal-to-noise ratio (SNR) using finite element (FE) and experimental studies. The study is concerned with part-circumferential part-depth planar cracks and the fundamental shear horizontal SH0 guided wave mode is considered. It is shown that PWI achieves superior resolution compared to CSM and comparable resolution to TFM. The techniques involving plane wave acquisition (PWI and CSM) are found to substantially outperform methods based on full matrix capture (FMC) in terms of SNR. Therefore, it is concluded that PWI which achieves good resolution and high SNR is a more attractive choice for short-range pipe guided wave testing, compared to other considered techniques. It was demonstrated that it is possible to estimate crack through wall thickness depth from the indication amplitude measured from the PWI reconstructed images. Subsequently, a novel PWI transduction setup is proposed, and it is shown to suppress the transmission of the unwanted fundamental S0 mode, which further improves SNR of PWI. It is often the case that access to the full pipe circumference is limited in pipelines with high feature density. Therefore, the influence of limited circumferential access on the performance of SH0 guided wave PWI is investigated. A decrease in circumferential extent is analytically shown to reduce the k-space data available for imaging and impair the phased array focusing performance. This is partially balanced by a temporal bandwidth of a typical probing Hann-windowed excitation which populates some of the missing parts of the k-space. The FE and experimental studies compare the performance of PWI for a range of different array circumferential extents. The key conclusions are that array circumferential extents above 60 % are required for the reliable detection, location, and length estimation of small cracks, 50 % circumferential extent only provides sufficient detection and location of cracks above a wavelength in length and array circumferential extents below 50 % are not suitable for reliable pipe inspection.