CAJ | 학술논문

为了实现对重庆市轻轨轨道梁的在役锚固螺杆的健康检测并且能够及时发现故障螺杆，本文提出了一种基于集合经验模态分解(EEMD)的样本熵与支持向量数据描述(SVDD)相结合的诊断方法。首先，利用 EEMD 方法将得到的正常锚固螺杆振动信号分解为若干个平稳的本征模态函数(IMF)分量，然后计算得到每个分量的样本熵，最后把得到的样本熵作为特征集输入到SVDD分类器进行训练和分类。大量实验结果表明，该方法不仅解决了故障样本缺乏的难题而且对锚固螺杆故障识别的错分率可以控制在5%以下，因此可以作为锚固螺杆是否处于正常工作状态的重要检测方法。
위료실현대중경시경궤궤도량적재역묘고라간적건강검측병차능구급시발현고장라간，본문제출료일충기우집합경험모태분해(EEMD)적양본적여지지향량수거묘술(SVDD)상결합적진단방법。수선，이용 EEMD 방법장득도적정상묘고라간진동신호분해위약간개평은적본정모태함수(IMF)분량，연후계산득도매개분량적양본적，최후파득도적양본적작위특정집수입도SVDD분류기진행훈련화분류。대량실험결과표명，해방법불부해결료고장양본결핍적난제이차대묘고라간고장식별적착분솔가이공제재5%이하，인차가이작위묘고라간시부처우정상공작상태적중요검측방법。
In order to achieve the health detection of the light-rail’s in-service screws in Chongqing and find fault screw in time, in this paper a method based on EEMD sample entropy and SVDD is proposed.First, decompose the normal screws’ signal into a number of stationary IMFS, then calculate the sample entropy of every IMF, in the end, put the sample entropy into SVDD classifier to be trained and classified.A large number of experimental results show that this method not only can solve the problem of the missing of the fault samples, but also can control the error rate below 5%, so this method is a good way to diagnosis the screws.