Radio frequency non-destructive testing and evaluation of defects under insulation

PhD ThesisThe use of insulation such as paint coatings has grown rapidly over the past decades. However, defects and corrosion under insulation (CUI) still present challenges for current non-destructive testing and evaluation (NDT&E) techniques. One of such challenges is the large lift-off introduced by thick insulation layer. Inaccessibility due to insulation leads corrosion and defects to be undetected, which can lead to catastrophic failure. Furthermore, lift-off effects due to the insulation layers reduce the sensitivities. The limitations of existing NDT&E techniques heighten the need for novel approaches to the characterisation of corrosion and defects under insulation. This research project is conducted in collaboration with International Paint®, and a radio frequency non-destructive evaluation for monitoring structural condition is proposed. High frequency (HF) passive RFID in conjunction with microwave NDT is proposed for monitoring and imaging under insulation. The small-size, battery-free and cost-efficient nature of RFID makes it attractive for long-term condition monitoring. To overcome the limitations of RFID-based sensing system such as effective monitoring area and lift-off tolerance, microwave NDT is proposed for the imaging of larger areas under thick insulation layers. Experimental studies are carried out in conjunction with specially designed mild steel sample sets to demonstrate the detection capabilities of the proposed systems. The contributions of this research can be summarised as follows. Corrosion detection using HF passive RFID-based sensing and microwave NDT is demonstrated in experimental feasibility studies considering variance in surface roughness, conductivity and permeability. The lift-off effects introduced by insulation layers are reduced by applying feature extraction with principal component analysis and non-negative matrix factorisation. The problem of thick insulation layers is overcome by employing a linear sweep frequency with PCA to improve the sensitivity and resolution of microwave NDT-based imaging. Finally, the merits of microwave NDT are identified for imaging defects under thick insulation in a realistic test scenario. In conclusion, HF passive RFID can be adapted for long term corrosion monitoring of steel under insulation, but sensing area and lift-off tolerance are limited. In contrast, the microwave NDT&E has shown greater potential and capability for monitoring corrosion and defects under insulation