【作者】 叶庆卫； 汪同庆；

【Author】 Ye Qingwei~(1,2) Wang Tongqing~1 (1 key Lab of Opto-electronic Tech.Under the Ministry of Education Chongqing University Chongqing 400044 China; 2 Information Science and Engineer College of Ningbo University Ningbo Zhejiang 315211 China)

【机构】 重庆大学光电技术及系统教育部重点实验室； 宁波大学信息科学与工程学院；

【摘要】 提出采用时频熵特征对在役锚固螺杆预紧力矩变化情况进行无损检测。依据轻轨锚固螺杆系统的单脉冲激励的物理振动特性,采用了相干项相对较少的伪魏格纳分布进行时频特性分析。在伪魏格纳分布图上提出了时间轴和频率轴投影曲线的信息熵特征来进行预紧力矩状态的检测。经过实际的振动数据处理计算发现,这种投影曲线的信息熵特征在不同螺杆的不同环境因素下能起到很好的预紧力矩故障检测和诊断作用。通过与采用其它特征的分类诊断比较发现,时频熵特征具有更好的分类性能。更多还原

【Abstract】 The vibration signals are tansmit into time-frequency distribution with pseudo-Wigner-Ville transformation.Base on the images of time-frequecncy distribution a kind of time entropy is defined as the information entropy of the time projection curve of the image and a kind of frequency entropy is defined as the frequency projection curve of the image.In fault detection experiment of light-rail bolt,these two entropy features are used to judge that the light-rial bolt is loosen or tighten Compared to other features like as time moment and frequency moment,the time-frequency entropy features work better in fault detection of light-rail bolt.更多还原