Damage detection method of CFRP structure based on fiber Bragg grating and principal component analysis

Damage detection method of CFRP structure based on fiber Bragg grating and principal component analysis
复制标题

基于光纤布拉格光栅和主成分分析的CFRP结构损伤检测方法

DOI:
10.1016/j.ijleo.2018.10.055
复制
发表时间:
2019-01-01
期刊:
影响因子:
3.1
通讯作者:
Yu, Hongliang
Yu, Hongliang
中科院分区:
物理与天体物理3区
文献类型:
--
作者:
Lu, Shizeng;Jiang, Mingshun;Yu, Hongliang

文献摘要

被引文献

相似文献

损伤检测是保证碳纤维增强塑料(CFRP)结构安全的重要环节。提出了一种基于光纤布拉格光栅传感器和主成分分析的复合材料结构损伤检测方法。首先,对未损伤的CFRP结构进行主动激励,利用FBG传感器监测其动态响应信号。将FBG传感器粘贴在CFRP结构的表面。然后,利用傅里叶变换方法提取频率响应作为损伤特征。然后,利用主成分分析对损伤特征进行降维。建立了仅基于非损伤状态样本的非损伤状态主成分模型,并以重构误差阈值作为损伤检测的指标。因此,可以通过确定新样本的重建误差是否大于重建误差阈值来检测结构损伤。最后,构建了CFRP结构损伤检测系统,验证了该方法的有效性。实验结果表明,本文提出的方法只需基于非损伤状态样本即可准确识别结构是否存在损伤。为CFRP结构的损伤检测提供了一种可行的方法。
Damage detection is very important to ensure the safety of carbon fiber reinforced plastics (CFRP) structure. In this paper, a damage detection method of composite structure based on fiber Bragg grating sensors and principal component analysis was proposed. Firstly, the non-damaged CFRP structure was excited by active excitation, and its dynamic response signal was monitored by FBG sensors. The FBG sensors were attached to the surface of the CFRP structure. And then, the frequency response was extracted as the damage feature by using the Fourier transform method. After that, principal component analysis was used to reduce the dimension of damage feature. The principal component model of the non-damaged state only based on the non-damaged state samples was established and the reconstruction error threshold was determined as the index of damage detection. Thus, structural damage can be detected by determining if the reconstruction error of the new sample was greater than the reconstruction error threshold. Finally, the damage detection system of CFRP structure was constructed to validate the proposed method. The experimental results showed that the method proposed in this paper can accurately identify whether the structure was damaged only based on the non-damaged state samples. This paper provided a feasible method for the damage detection of CFRP structure.