Noncontact visualization of multiscale defects in CFRP composites using eddy current testing with T-R probe

Abstract

Carbon fiber reinforced polymer (CFRP) is widely used in modern industries, and damages may occur from manufacturing to utilization, highlighting the importance of CFRP detection. As being a multi-phase structural material, the CFRP's mesostructural features contained in eddy current signals can obscure macroscopic defect characteristics. Phase rotation is a common method for eddy current signal processing, but the specific criteria for evaluation of processing effect are lacking. This paper proposes three reliable criteria for assessment of the phase rotation processing effects of Transmitter-Receiver (T-R) probe signals, namely: signal overlap ratio (SOR), signal-to-noise ratio (SNR), and image average gradient (IAG). To begin, the SOR serves as a crucial indicator for assessing signal similarity and the degree of overlap, with a low SOR indicating the method's effectiveness in distinguishing defect characteristics. Subsequently, by comparing the SNRs of images, the clarity of defects under different rotation angles can be effectively quantified. Finally, a novel criterion, i.e. the IAG, is introduced. What distinguishes this criterion is its higher degree of automation, facilitating a comprehensive evaluation of processing effects without excessive subjective intervention. After analyzing these criteria, it is clear that the IAG criterion excels in automation and objectivity for evaluating eddy current signal processing effects.