转子—滚动轴承耦合系统的滚动轴承故障分析与智能诊断

发布时间：2018-03-15 02:09

本文选题：滚动轴承　切入点：小波变换　出处：《南京航空航天大学》2011年硕士论文　论文类型：学位论文

【摘要】：滚动轴承是航空发动机重要的支承部件,它的运行状态是否正常直接影响到整台发动机的精度、可靠性及寿命等。因此,深入研究滚动轴承故障机理和振动特性,准确诊断轴承故障具有重要意义。本文以真实的转子-滚动轴承试验台为研究对象,建立了含轴承故障的转子-滚动轴承耦合动力学模型,研究轴承故障机理,提取了轴承故障特征,并提出了一种滚动轴承故障诊断新方法。具体研究内容如下: 第一、本文以真实试验台为原型,建立了转子-滚动轴承模型,该模型将转子考虑为两端无约束的等截面自由欧拉梁模型,并对滚动轴承常见故障进行了详细建模,考虑了滚动轴承的非线性接触力,最后运用模态截断法,将系统偏微分方程转化为常微分方程组,并运用新型显式数值积分方法获取转子-轴承系统的响应。同时对真实试验台进行了模态实验分析,相同参数条件下,实验模态与计算模态基本一致,验证了模型的正确有效性;仿真分析了滚动轴承内、外圈和滚珠故障时振动信号,分析结果与理论结果一致。第二、基于小波变换、包络解调及盲源分离技术,研究了基于独立分量分析的滚动轴承振动信号降噪方法,通过实测信号验证,表明了基于ICA降噪方法能有效地提高监测信号的信噪比。第三、提取滚动轴承振动信号中4个无量纲的时域参数和频域中小波包络特征参数,运用数据挖掘平台Weka软件,借助Weka平台的C4.5决策树提取了滚动轴承故障知识规则,并对规则加以解释,最后应用该知识规则诊断从试验器上模拟的大量轴承数据,诊断结果表明了该知识规则具有较高的正确性和实用性。第四、建立基于LabVIEW的滚动轴承数据采集系统和智能诊断系统,利用数据挖掘平台Weka软件获取的滚动轴承故障知识规则对滚动轴承进行了智能诊断。
[Abstract]:Rolling bearing is an important supporting part of aero-engine. Whether its running state is normal or not directly affects the accuracy, reliability and life of the whole engine. Therefore, the fault mechanism and vibration characteristics of rolling bearing are deeply studied. It is of great significance to diagnose bearing fault accurately. In this paper, a rotor-rolling bearing coupling dynamic model with bearing fault is established, and the bearing fault mechanism is studied by taking the real rotor-rolling bearing test rig as the research object. A new fault diagnosis method for rolling bearing is proposed. The main contents are as follows:. Firstly, the model of rotor-rolling bearing is established based on the real test bed. The model considers the rotor as a free Euler beam model with unconstrained ends, and the common faults of rolling bearing are modeled in detail. Considering the nonlinear contact force of rolling bearings, the partial differential equations of the system are transformed into ordinary differential equations by modal truncation method. The response of rotor-bearing system is obtained by using a new explicit numerical integration method, and the modal analysis of the real test rig is carried out. Under the same parameter condition, the experimental mode is basically the same as the calculated mode. The validity of the model is verified, and the vibration signals of the inner, outer ring and ball are simulated and analyzed, and the results are in agreement with the theoretical results. Secondly, based on wavelet transform, envelope demodulation and blind source separation, the noise reduction method of rolling bearing vibration signal based on independent component analysis is studied. It shows that the noise reduction method based on ICA can effectively improve the signal-to-noise ratio of the monitored signal. Thirdly, four dimensionless time-domain parameters and frequency-domain wavelet envelope characteristic parameters are extracted from the vibration signals of rolling bearings. The fault knowledge rules of rolling bearings are extracted by using Weka software of data mining platform and C4.5 decision tree of Weka platform. The rules are explained and applied to diagnose a large number of bearing data simulated from the tester. The diagnostic results show that the knowledge rules are accurate and practical. In 4th, a rolling bearing data acquisition system and an intelligent diagnosis system based on LabVIEW were established. The rolling bearing fault knowledge rules obtained by the data mining platform Weka software were used to diagnose the rolling bearing intelligently.
【学位授予单位】：南京航空航天大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：TH165.3

【相似文献】