Start Date

2016 12:00 AM

Description

Guided wave measurements are an established inspection method for pipelines in a variety of industries. Permanently installed guided wave sensors (gPIMS) have been used in the past to reduce the access cost for pipes in hard to reach locations which require repeated inspections. Making use of the existing gPIMS technology a method was devised to move from the current single inspection approach to a structural health monitoring system. With changes due to the environmental conditions of the pipe, such as temperature variations or content changes, the usefulness of simple baseline subtraction methods is very limited as temperature compensation methods do not allow for the full removal of these effects. Using a dataset of historic measurements and statistical signal processing it was possible to identify and track defects in pipe sections. The procedure is based on Independent Component Analysis (ICA) and allows the collected measurements to be separated into components of the constant baseline and those components associated with signal changes. Further processing of these components allows the removal of those caused by environmental variations or noise, leaving only components related to real changes of the pipe structure.

Datasets were obtained from two different field trials, one of which was open, with known defect locations and positions, while the other was a blind trial. It was found that with the presented method it was possible to detect and track the growth of defects in the pipe structure with reflection coefficients of down to 0.5%. Defects at welds and on pipe bends were also detected and tracked successfully. Structural changes of this order cannot be detected reliably using standard guided wave processing methods so this new approach promises a significant advance in monitoring performance.

Language

en

File Format

application/pdf

Share

COinS
 
Jan 1st, 12:00 AM

Structural Health Monitoring of Pipelines Using Permanently Installed Guided Wave Sensors

Guided wave measurements are an established inspection method for pipelines in a variety of industries. Permanently installed guided wave sensors (gPIMS) have been used in the past to reduce the access cost for pipes in hard to reach locations which require repeated inspections. Making use of the existing gPIMS technology a method was devised to move from the current single inspection approach to a structural health monitoring system. With changes due to the environmental conditions of the pipe, such as temperature variations or content changes, the usefulness of simple baseline subtraction methods is very limited as temperature compensation methods do not allow for the full removal of these effects. Using a dataset of historic measurements and statistical signal processing it was possible to identify and track defects in pipe sections. The procedure is based on Independent Component Analysis (ICA) and allows the collected measurements to be separated into components of the constant baseline and those components associated with signal changes. Further processing of these components allows the removal of those caused by environmental variations or noise, leaving only components related to real changes of the pipe structure.

Datasets were obtained from two different field trials, one of which was open, with known defect locations and positions, while the other was a blind trial. It was found that with the presented method it was possible to detect and track the growth of defects in the pipe structure with reflection coefficients of down to 0.5%. Defects at welds and on pipe bends were also detected and tracked successfully. Structural changes of this order cannot be detected reliably using standard guided wave processing methods so this new approach promises a significant advance in monitoring performance.