Damage detection of carbon fiber reinforced polymer composite materials based on one-dimensional multi-scale residual convolution neural network

Abstract

Carbon fiber reinforced polymers (CFRPs) have been widely applied in the aerospace industry, and the health conditions of CFRPs largely affect aerospace safety. Due to the limitations of traditional detection methods, electrical impedance tomography (EIT) has been gradually applied in the damage detection of CFRP composite materials. Aiming at the problems of poor imaging quality and low identification rate in the traditional EIT reconstruction algorithm, an EIT algorithm based on the one-dimensional multi-scale residual convolution neural network (1D-MSK-ResNet) is proposed in this paper. A “voltage vector-conductivity media distribution” dataset is first established, and the training results of the testing dataset are used to verify and evaluate the algorithm. Simulation and experimental results indicated that the 1D-MSK-ResNet EIT algorithm could enhance the ability of damage identification and significantly improve the imaging quality.

Damage identification of wire rope under noise background via Light- EfficientNetV2 and Magnetic flux leakage image

“…CNN in machine learning has fast speed, high accuracy, and good robustness for image classi cation [17][18][19], so it is widely used in fault identi cation and classi cation [20][21][22]. Some scholars use the characteristics of CNN to convert the damage signals in the fault eld into images and input them into CNN for damage identi cation.…”