节点文献

基于一类支持向量机的管道安装应力智能检测技术

Intelligent Detection Technology of Pipe Installation Stress Based on One-class Support Vector Machine

  • 推荐 CAJ下载
  • PDF下载
  • 不支持迅雷等下载工具,请取消加速工具后下载。

【作者】 於为刚陈果寸文渊赵正大张凯勇陈雪梅侯民利

【Author】 YU Wei-gang;CHEN Guo;CUN Wen-yuan;ZHAO Zheng-da;ZHANG Kai-yong;CHEN Xue-mei;HOU Min-li;College of Civil Aviation, Nanjing University of Aeronautics and Astronautics;Chengdu Aircraft Industrial (Group) Co., Ltd.;Engineering Training Center, Nanjing University of Aeronautics and Astronautics;

【机构】 南京航空航天大学民航学院成都飞机工业(集团)有限责任公司南京航空航天大学工程训练中心

【摘要】 基于实验研究了初始安装应力对管道振动固有特性的影响规律,提出了基于一类支持向量机的管道安装应力智能检测技术。首先构建管道安装应力检测试验台,模拟了初始安装应力,利用锤击法对管道进行模态试验,得到了安装应力对管道固有特性的影响规律;在此基础上,提出了基于一类支持向量机的管道安装应力智能检测技术,并利用试验台安装应力试验数据进行方法验证。试验结果表明:随着装配应力增加,检测结果偏离正常状态的程度也逐步增加,充分验证了该检测技术对于管道安装应力检测的灵敏性和有效性。

【Abstract】 Based on the experimental study of the influence laws of initial installation stress on the inherent vibration characteristics of pipelines, an intelligent detection technology of pipeline installation stress based on one-class support vector machines was proposed. Firstly, a pipeline installation stress test-bed was constructed, which simulated the initial installation stress. The influence laws of the installation stresses on the inherent characteristics of the pipeline were obtained by modal tests conducted on the pipeline through the hammer method. On this basis, a kind of intelligent detection technology of pipeline installation stress based on one-class support vector machine was proposed, and the installation stress test data of the test-bed was used for method validation. Experimental results show that the deviation of test results from the normal state is gradually increased with the increase of installation stress, which fully verifies the sensitivity and effectiveness of this detection technology for pipeline installation stress test.

【基金】 国家自然科学基金资助项目(51675263);成都飞机工业(集团)有限责任公司资助项目
  • 【文献出处】 管道技术与设备 ,Pipeline Technique and Equipment , 编辑部邮箱 ,2019年02期
  • 【分类号】V263.2
  • 【被引频次】4
  • 【下载频次】124
节点文献中: 

本文链接的文献网络图示:

本文的引文网络