http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
음향방출시험(AET) 기반 CFRP 복합소재 결함 분류 딥러닝 모델 개발
김호연(Ho-Yeon Kim),김경영(Gyeong-Yeung Kim),김다현(Da-Hyun Kim),김동주(Dong-Ju Kim) 대한전자공학회 2022 대한전자공학회 학술대회 Vol.2022 No.11
CFRP, which consists of composite materials and can easily be exposed to a high-pressure environment, requires a systematic safety inspection to identify specific defects and failure types. AET is a suitable method for fault diagnosis because AE signals can be collected sensitively to the source of defects. However, there is no system to accurately diagnose the fault of CFRP. This study introduces a method for identifying defects in CFRP using the deep learning-based model of AE signal. The dataset was established by collecting defective sensor waveforms generated during the tensile test on CFRP materials. A deep learning-based 1DCNN model was developed by classifying each source, showing 91.6% accuracy. Through this study, a prompt and accurate diagnosis of CFRP can be established resulting in structural stability.