Evaluation of Signal Post Processing to Improve Interpretation of Guided Wave Testing on Pipes with Attenuating Coating
By Kamran Pedram and Peter Mudge
Industrial Need
The performance of guided wave testing on coated and buried pipelines, especially the test range and defect sensitivity achievable, is of particular importance as such pipelines are generally only accessible for testing from a limited number of locations, which may need to be specially prepared by excavation. However, where pipes have a protective coating, the coating material is normally viscoelastic and absorbs sound energy and, if the pipeline is buried, sound energy can also interact with the surrounding medium, causing further losses. In both cases, attenuation of guided waves is high (of the order of several dB per metre) in the usual frequency range for long range ultrasonic testing from 20kHz to 100kHz, so that the effective test range is significantly reduced. Furthermore, in addition to the reduction in signal strength, the background noise level can also increase, especially for coating materials that have been present on the pipe for a number of years. In this case, bonding conditions between the coating material and the pipe substrate become uneven due to changes to the coating properties over time, pressure changes and temperature variations around the pipe etc, causing noise signals to be created by scattering and mode conversion owing to the varying interface conditions. This has an impact on the sensitivity of the test to defects compared with guided wave tests on uncoated pipe. To maintain sensitivity, it is necessary to identify small signals that may be within the noise floor, and signal interpretation is very challenging because of the complexity of the noise signature.
TWI has been developing post-processing methods based on split-spectrum processing (SSP) in order to reduce the noise on the guided wave signals caused by such scattering. The purpose of this report is to present the findings of a validation exercise, which has used data from earlier guided wave tests on highly attenuating pipes, containing deliberately introduced defects, to assess the ability of SSP to identify defects in coated and buried pipes which are not detectable using standard guided wave techniques.
Key findings
- The use of SSP allowed a clearer distinction between signals from real reflectors and other clutter on the signal and reduced the number of false calls made by the interpreter.
- The detection performance was not improved by the application of SSP.
- Application of SSP did not result in an increase of effective test range on highly attenuating pipes.