旋转机械轴承振动信号分析方法研究

发布时间：2018-01-10 03:27

本文关键词：旋转机械轴承振动信号分析方法研究　出处：《重庆大学》2014年博士论文　论文类型：学位论文

【摘要】：轴承作为旋转机械中广泛使用的关键零部件之一，其运行状态直接关系整台机械设备的工作性能，开展基于振动信号分析的轴承状态监测与故障诊断的相关研究并及时准确地识别故障萌发与演变，对确保设备平稳运行、减少甚至避免重大安全事故具有相当重要的意义。本课题以滚动轴承及滑动轴承故障振动信号为研究对象，针对滚动轴承故障信号易受噪声干扰的影响拓展并丰富了峭度图理论在振动信号降噪以及故障特征提取中的应用，并基于定向循环平稳分析理论研究了滑动轴承在油膜失稳状态下的故障信号特征，，最后开发了虚拟式旋转机械轴承测试诊断系统实现了理论创新成果在工程实践中的应用。本文的具体研究内容介绍如下：本文首先介绍了峭度图理论中涉及的峭度统计理论、谱峭度系数等数学基础，并详细阐述了传统峭度图算法以及基于COT的阶比峭度图算法在故障信号降噪和最优解调频带参数确定方面的优势。针对传统四阶矩累积量谱峭度系数易受信号奇异点影响而不能真实估计信号峰态程度水平的问题，定义了Moors谱峭度、Hogg谱峭度以及Crow-Siddiqui谱峭度等三种鲁棒性谱峭度系数，并提出了能消除信号奇异点干扰的鲁棒性峭度图算法。滚动轴承仿真及实测故障信号分析结果验证了鲁棒性峭度图与传统峭度图相比，能显著提高滤波效果并增强原始信号中的故障瞬态冲击特征。传统的基于STFT、FIR滤波器以及第一代小波包变换的峭度图算法中存在时频分辨率取舍、运算效率以及小波函数库有限等缺点，本文提出了基于尺度自适应冗余提升小波包变换的峭度图算法，通过基于原始信号本身构造更新器和预测器来实现信号分解与重构的自适应功能，有效地提高了峭度图算法的自适应特性以及滤波后故障信号的准确性。随后利用时间尺度分解进一步增强了峭度图滤波后信号的瞬态冲击故障特征，并将基于固有时间尺度分解的Hilbert谱作为包络解调分析的有益补充，通过与Hilbert-Huang变换的对比验证了其在表征故障信号时频能量信息分布中的优越性。旋转机械振动信号往往呈现循环平稳特性，本文将机械故障诊断领域中用于分析单通道实信号的循环平稳统计量理论进行了拓展，定义了能用于分析双通道融合复信号的低阶定向循环平稳统计量并提取出了滑动轴承油膜失稳故障信号中周期性时变特征以及转子振动状态信息等。通过对比分析揭示了传统全谱分析以及定向Wigner分布与定向循环统计量之间的内在联系，即全谱分析实质上就是一阶定向循环平稳统计量（定向循环均值），而定向Wigner分布实质上属于二阶定向循环统计量（定向循环谱相关函数）。此外试验结果分别对比分析了加速度振动信号与位移振动信号中所包含系统振动状态信息。根据轴承振动监测需求分析，基于NI Labview平台开发了轴承振动信号测试、分析与特征提取于一体的虚拟式旋转机械轴承测试系统，为取得的理论创新成果在工程实践中的应用提供了强有力的支持。文章最后对全文工作及主要创新点进行了总结，并展望了后续的研究方向。
[Abstract]:The bearing is one of the key parts widely used in rotary machinery . The running state of the bearing is directly related to the working performance of the whole mechanical equipment , the research on the bearing state monitoring and fault diagnosis based on the vibration signal analysis is carried out , and the fault signal characteristics of the sliding bearing under the unstable state of the oil film are analyzed . In this paper , we first introduce the mathematical base of the theory , the coefficient of the spectrum , and so on , which are involved in the theory of Kurtosis graph , and give a detailed description of the advantages of the traditional Kurtosis graph algorithm and the optimal demodulation frequency band parameter determination based on the first order - rate graph algorithm . The robustness of the Moors ' spectrum , Hogg ' s spectrum and the degree of the degree of the optimal demodulation band are defined . The simulation of the rolling bearing and the analysis of the measured fault signal show that the robustness - ness graph can obviously improve the filtering effect and enhance the fault transient impact characteristics in the original signal . The traditional STFT , FIR filter and the first generation wavelet packet transform have the disadvantages of time - frequency resolution acquisition , operation efficiency and limited wavelet function library . Based on the original signal itself , the adaptive function of signal decomposition and reconstruction is proposed . By using time - scale decomposition , the transient impact fault characteristics of the signal are enhanced . The Hilbert spectrum decomposed by the inherent time scale is used as the beneficial complement of the envelope demodulation analysis . By comparison with Hilbert - Huang transform , the superiority of this algorithm in characterizing the frequency energy information distribution of the fault signal is verified . In this paper , the theory of cyclic stationary statistics for the analysis of single channel real signals is expanded in the field of mechanical fault diagnosis , and the relationship between the periodic time - varying feature and the rotor vibration state information is defined in the field of mechanical fault diagnosis . According to the analysis of bearing vibration monitoring demand , the test system of bearing vibration signal is developed based on NI Labview platform , and the virtual rotary mechanical bearing test system is developed , which provides powerful support for the application of the theoretical innovation results in engineering practice . Finally , the paper summarizes the whole work and the main innovation points , and looks forward to the future research direction .

【学位授予单位】：重庆大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TH133.3;TH165.3

【参考文献】