基于光纤光栅与BP神经网络的孔边裂纹监测研究

Research on hole edge crack monitoring based on optical fiber gratings and BP neural network

含孔金属结构的孔边裂纹监测对于保障飞行安全,增强飞机结构可靠性具有重要意义。为实现对孔边裂纹扩展的监测,进行含有孔边角裂纹的含孔铝合金板疲劳加载试验,得到含孔铝合金板试验件的a-N曲线以及孔边裂纹扩展过程中光纤光栅应变传感器中心波长偏移量;利用包络分析法、BP神经网络等损伤识别算法对试验数据进行处理与分析;建立能够以光纤光栅应变传感器中心波长偏移量识别孔边裂纹扩展的监测模型,并通过试验对监测模型进行验证。结果表明:本文建立的监测模型能够有效识别出孔边角裂纹的扩展与穿透,对孔边角裂纹扩展长度监测的准确度达到了97.2%,未来可应用于全机地面疲劳试验、飞机结构健康监测等多种场景。

The hole edge crack monitoring of metal structures with holes is of great significance for ensuring flight safety and enhancing the reliability of aircraft structures.In order to monitor the crack growth at the hole edge,the fatigue loading test of porous aluminum alloy plate containing the corner crack at the hole edge is carried out,and the a-N curve of the test piece of porous aluminum alloy plate and the center wavelength offset of the optical fiber grating strain sensor during the crack growth at the hole edge are obtained.The damage identification algorithms such as envelope analysis method and BP neural network are used to process and analyze the test data.The monito-ring model that can identify the crack growth at the hole edge with the center wavelength offset of the optical fiber grating strain sensor is established,and verified with test parts.The results show that the established monitoring model can effectively identify the propagation and penetration of the corner crack at the hole edge,and the accuracy of monitoring the propagation length of the corner crack at the hole edge has reached 97.2%,which can be applied to the ground fatigue test of the whole aircraft,aircraft structure health monitoring and other scenarios in the future.