Bolt Looseness Detection under Changing Temperature Conditions using Independent Component Analysis.
Abstract »Abstract: Damage detection technology based on guided wave is used widely in structural health monitoring, which relies on comparing response signals with baseline records. However, the temperature variation can also produce large changes in guided wave signals, thereby causing a false alarms of damage detection. In this paper, independent component analysis (ICA) method was developed to process guided wave response signal for separating damage-induced information from coherent interference signal. As a result, the influence of temperature variation could be eliminated in the process of damage localization. In the laboratory, the experiment of bolt looseness damage detection was made in the environment of simulated temperature variation. The collected signals were utilized in damage location estimation after processing by ICA. It showed good result for bolted looseness localization, which indicated that this method is robust to the temperature variation exhibited in damage detection based on guided wave.