CAJ | 학술논문

运用神经网络技术，设计开发了一种基于虚拟仪器平台的轴承故障检测系统。在LabVIEW软件平台上，分别对轴承正常和待检测状态信号进行时域分析并提取其特征值，创建神经网络的训练样本和被测样本，并代入MATLAB节点中的神经网络进行故障诊断。试验结果表明，该系统能正确、快速地判断出被测的轴承是否正常，识别率达100％，当被测轴承诊断出故障时，系统还可发出报警提示。
운용신경망락기술，설계개발료일충기우허의의기평태적축승고장검측계통。재LabVIEW연건평태상，분별대축승정상화대검측상태신호진행시역분석병제취기특정치，창건신경망락적훈련양본화피측양본，병대입MATLAB절점중적신경망락진행고장진단。시험결과표명，해계통능정학、쾌속지판단출피측적축승시부정상，식별솔체100％，당피측축승진단출고장시，계통환가발출보경제시。
A fault detection system for bearings based on virtual instrument platform is designed and developed by using neural network technology.The system is developed on LabVIEW software platform,in which both the normal state sig-nal and the signal to be detected are analyzed in time-domain to extract the feature values,and create the trained and tested samples of neural network,and then the trained and tested samples are put into neural network of MATLAB node to realize fault diagnosis.The experimental results show that the system is able to judge whether the bearing state is nor-mal or not accurately and quickly,with a 100% recognition rate,and the system alarms and prompts when a fault oc-curs.