毕业论文 基于参数化模型的滚动轴承特征提取及其智能诊断研究.docVIP

PAGE \* MERGEFORMAT PAGE \* MERGEFORMAT 2 PAGE \* MERGEFORMAT I PAGE \* MERGEFORMAT I PAGE \* MERGEFORMAT I 毕业论文 基于参数化模型的滚动轴承特征提取及其智能诊断研究 摘要 随着机械设备向复杂化、自动化方向的高速发展，滚动轴承广泛运用于各种旋转机械设备，是旋转设备中非常关键的部件。从而对滚动轴承的智能诊断在现代生产中起着越来越大的作用。为了获得更高的故障识别率，基于信号处理与神经网络相结合实现故障智能诊断成为了当今主要的发展方向。在复杂工况条件下，滚动轴承的振动信号显示出非平稳性，因此对非平稳信号的特征提取并对其故障进行智能识别是非常必要的。 本文从滚动轴承振动信号的采集入手，对采集的非平稳信号进行参数化建模，本文使用了TVAR建模分析方法，基于基函数算法求解该模型的时变参数，采用Akaike Information Criterion(AIC)准则确定模型阶数。同时，对TVAR模型难以跟踪突变信号的问题，本文研究了利用Garch建模分析的方法，通过仿真信号分析表明Garch的条件方差能很好的反映信号频率的突变性或冲击性。在利用上述参数化模型对振动信号进行特征提取的基础上，构建BP神经网络，对滚动轴承的不同故障进行智能诊断。 在上述研究基础上，基于 LabVIEW 开发了滚动轴承振动信号的采集和参数化模型分析以及智能诊断软件。实现了滚动轴承振动的非平稳信号的采集、时频分析以及基于TVAR、Garch与神经网络的智能诊断、查询与保存诊断记录等功能。这些研究成果对滚动轴承振动信号特征的提取及智能诊断具有重要意义。 关键词：滚动轴承；非平稳信号；时变自回归；广义条件异方差；特征提取；智能诊断 PAGE \* MERGEFORMAT 4 PAGE \* MERGEFORMAT 4 PAGE \* MERGEFORMAT 1 PAGE \* MERGEFORMAT 1 PAGE \* MERGEFORMAT 12 Study on Feature Extraction of Rolling Bearing and Intelligent Diagnosis Based on Parametric Model ABSTRACT As the complicated and automatic development of machine,Rolling bearings are widely used in various rotating machinery equipment,and are very critical in the rotating equipment parts.So there is more and more effect of intelligent diagnosis in modern manufacturing industry. In order to get higher Fault identification ratio, the fault diagnosis technique based on the combination of signal processing and neural network model has become the host research field. Actually, the vibration signals of mechanical equipments have nonstationary characteristics under complex working condition. Therefore, it is very important to develop the research on features extraction of nonstationary signals and automatic fault identification. Start with acquisition of rolling bearing vibration signal, Parameterized modeling of nonstationary signals, the TVAR model was used, The time varying parameters was solved by basis function arithmetic respectively ,And the Akaike Information Criterion(AIC) ru