| 陈振华,涂思敏,殷晓芳,邓文武,张志强,卢超.航空蜂窝夹芯复合材料板粘接状态智能超声检测技术[J].电子测量与仪器学报,2026,40(2):303-310 |
| 航空蜂窝夹芯复合材料板粘接状态智能超声检测技术 |
| Intelligent ultrasonic testing technology for bonding status ofaerial honeycomb sandwich composite plate |
| |
| DOI： |
| 中文关键词: 蜂窝夹芯复合板 喷水式超声聚焦检测 深度学习 智能分类 粘接状态 C扫描图像 |
| 英文关键词:honeycomb sandwich composite plate water-jet ultrasonic focus testing deep learning intelligent classification bonding status C scan images |
| 基金项目:国家自然科学基金(12464059)、江西省重点研发计划项目(20212BBE51006)资助 |
|
|
| Author | Institution |
| Chen Zhenhua | Key Laboratory of Nondestructive Testing of Ministry of Education, Nanchang Hangkong University,
Nanchang 330063, China |
| Tu Simin | 1.Key Laboratory of Nondestructive Testing of Ministry of Education, Nanchang Hangkong University,
Nanchang 330063, China; 2.Jiangxi Hongdu Aviation Industry Group, Nanchang 330000, China |
| Yin Xiaofang | Jiangxi Hongdu Aviation Industry Group, Nanchang 330000, China |
| Deng Wenwu | Jiangxi Hongdu Aviation Industry Group, Nanchang 330000, China |
| Zhang Zhiqiang | Jiangxi Hongdu Aviation Industry Group, Nanchang 330000, China |
| Lu Chao | Key Laboratory of Nondestructive Testing of Ministry of Education, Nanchang Hangkong University,
Nanchang 330063, China |
|
| 摘要点击次数: 331 |
| 全文下载次数: 217 |
| 中文摘要: |
| 飞机平尾蜂窝夹芯复合板结构复杂、面积大、缺陷类型多，通过超声C扫描成像技术能够直观分析蜂窝与蒙皮粘接状态；由此产生大量的检测图像，需依赖技术人员丰富的工作经验进行评估，存在评估效率低、主观性强等问题。因此，提出基于深度学习的蜂窝夹芯复合材料板粘接层超声C扫描图像智能分类技术。首先，通过界面反射波跟踪方法采集胶层与蜂窝间界面的超声C扫描检测图像，结合图像处理技术进一步提高图像质量；其次，为构造训练数据库，通过滑动窗方式截取C扫描图像样本，根据C扫描幅度分布将样本分为3种粘接状态（目标区域），提出小样本图像数据集扩展方法；最后，构建50层残差网络（residual network 50 layers，ResNet50）并对其进行训练，并评估深度学习网络对蜂窝粘接状态的分类能力。研究结果表明，通过界面反射波跟踪能够克服蒙皮表面形状变化并形成蜂窝粘接层C扫描图像，ResNet50网络能够识别蜂窝夹芯复合板结构的3类目标区域，具有良好的稳定性和准确率，并体现出“智能性”特点。 |
| 英文摘要: |
| The aircraft flat-tail honeycomb sandwich composite plate has complex structure, large area and many types of defects. The bonding state between honeycomb and skin can be intuitively analyzed by ultrasonic C-scan imaging. As a result, a large number of detection images need to be evaluated by the rich work experience of technicians, there are problems such as low evaluation efficiency and strong subjectivity. Therefore, an intelligent classification technology of ultrasonic C-scan image of honeycomb sandwich composite plate bonding layer based on deep learning is proposed. Firstly, an ultrasonic C-scan testing image of the interface between the adhesive layer and the honeycomb is acquired by an interface reflected wave tracking method, and the image quality is further improved by combining an image processing technology. Secondly, in order to construct the training database, C-scan image samples are intercepted by sliding window, and the samples are divided into three bonding states (target areas) according to the amplitude distribution of C-scan, and a small sample image data set expansion method is proposed. Finally, the 50-layer residual network (ResNet50) is constructed and trained, and the classification ability of the deep learning network for honeycomb bonding states is evaluated. The results show that the interface reflection wave tracking can overcome the shape change of the skin surface and form the C-scan image of the honeycomb bonding layer, and the ResNet50 network can identify the three types of target areas of the honeycomb sandwich composite panel structure with good stability and accuracy, and reflect the characteristics of “intelligence”. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
