旋转机械全息谱分析精度提高方法的研究
发布时间：2019-01-23 11:59
【摘要】：全息谱技术是旋转机械振动信号分析与故障诊断中的一项重要技术。基于FFT的传统谱分析基本能够满足常见的故障特征提取和故障分析的要求,但其不能有效地利用振动信号的相位信息,因此不能满足精密诊断和复杂故障诊断的要求,全息谱技术正是在这一情况下发展起来的。全息谱技术的主导思想是充分利用传统谱分析所忽略相位信息,集成了一个或多个截面上水平和垂直通道的振动信号的幅值、频率和相位信息,用合成的一系列椭圆来描述不同频率分量下转子的振动行为,因此比一般方法更能准确识别和诊断旋转机械隐含的故障特征。本课题以旋转机械全息谱分析技术为研究对象,研究并提高了全息谱在密集频谱和噪声情况下的分析精度,跟踪并实现了谐波小波轴心轨迹提纯,最后进行了全息谱分析系统的开发。具体研究工作如下: ①在详细介绍全息谱技术测试原理的基础上,采用基于一阶数字微分和最小二乘五点三次平滑法来计算转速曲线,为准确获取工频提供保障。同时,通过分析比值校正法原理及具备更强抗噪能力的相位差校正法原理,发现现有的两种频域插值算法无法实现密集频谱校正,强调消除或弱化密集频谱及噪声对全息谱分析精度影响的必要性。 ②针对全息谱在密集频谱情况下精度降低的问题,研究并实现了基于频谱细化和相位差校正的全息谱。它是采用复解析带通滤波器选带细化方法对以阶次频率为中心的局部区间进行细化分析,通过提高频率分辨率从而消除密集频谱的影响,运用相位差法对细化区间内的最大谱峰进行频率、幅值、相位校正,准确地提取出各阶次的幅值、相位信息,最后合成全息谱图。仿真和现场实验研究表明,本文所研究的基于频谱细化和相位差校正的全息谱分析方法是正确而且有效的。 ③跟踪国内外轴心轨迹提纯研究研究状况,在综合分析现有轴心轨迹提纯方法优缺点的基础上,选用谐波小波提纯于本全息谱分析系统中,实现了算法过程,并用仿真实例验证了算法的正确性及有效性。 ④采用面向对象的编程技术以及VC++开发平台,开发了旋转机械全息谱分析系统,并利用实验数据对系统的各项功能模块进行了应用测试。测试效果表明本文所开发的旋转机械全息谱分析系统具有较好的工程应用价值
[Abstract]:Holographic spectrum is an important technique in vibration signal analysis and fault diagnosis of rotating machinery. The traditional spectrum analysis based on FFT can basically meet the requirements of common fault feature extraction and fault analysis, but it can not effectively utilize the phase information of vibration signal, so it can not meet the requirements of precision diagnosis and complex fault diagnosis. It is in this context that holographic spectroscopy is developed. The main idea of holographic spectrum technology is to make full use of the phase information neglected by traditional spectral analysis, and to integrate the amplitude, frequency and phase information of vibration signals of horizontal and vertical channels on one or more sections. A series of ellipses are used to describe the vibration behavior of the rotor with different frequency components, so it is more accurate to identify and diagnose the hidden fault characteristics of rotating machinery than the general method. In this paper, the holographic spectrum analysis technology of rotating machinery is taken as the research object, the analysis accuracy of the holographic spectrum under dense spectrum and noise is studied and improved, and the harmonics wavelet axis track is tracked and purified. Finally, the holographic spectrum analysis system is developed. The research work is as follows: 1 on the basis of the detailed introduction of the holographic spectrum testing principle, the rotational speed curve is calculated based on the first order digital differential and least square smoothing method, which provides the guarantee for the accurate acquisition of power frequency. At the same time, by analyzing the principle of ratio correction method and phase difference correction method with stronger anti-noise ability, it is found that two kinds of frequency domain interpolation algorithms can not achieve intensive spectrum correction. The necessity of eliminating or weakening the influence of dense spectrum and noise on the accuracy of holographic spectrum analysis is emphasized. 2 in order to reduce the precision of holographic spectrum in dense spectrum, the holographic spectrum based on spectrum refinement and phase difference correction is studied and realized. It uses the method of complex analytic bandpass filter to refine the local interval centered on the order frequency, and eliminates the influence of the dense frequency spectrum by improving the frequency resolution. The phase difference method is used to correct the frequency, amplitude and phase of the maximum spectral peak in the thinning region. The amplitude and phase information of each order are extracted accurately. Finally, the holographic spectrum is synthesized. Simulation and field experiments show that the holographic spectrum analysis method based on spectrum refinement and phase difference correction is correct and effective. (3) tracking the research status of axis trajectory purification at home and abroad, on the basis of synthetically analyzing the advantages and disadvantages of the existing methods, the harmonic wavelet is selected to purify the holographic spectrum analysis system, and the algorithm process is realized. A simulation example is given to verify the correctness and effectiveness of the algorithm. (4) the holographic spectrum analysis system of rotating machinery is developed by using object-oriented programming technology and VC development platform, and the functional modules of the system are tested with experimental data. The test results show that the holographic spectrum analysis system developed in this paper has good engineering application value.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：TH165.3
本文编号：2413744
[Abstract]:Holographic spectrum is an important technique in vibration signal analysis and fault diagnosis of rotating machinery. The traditional spectrum analysis based on FFT can basically meet the requirements of common fault feature extraction and fault analysis, but it can not effectively utilize the phase information of vibration signal, so it can not meet the requirements of precision diagnosis and complex fault diagnosis. It is in this context that holographic spectroscopy is developed. The main idea of holographic spectrum technology is to make full use of the phase information neglected by traditional spectral analysis, and to integrate the amplitude, frequency and phase information of vibration signals of horizontal and vertical channels on one or more sections. A series of ellipses are used to describe the vibration behavior of the rotor with different frequency components, so it is more accurate to identify and diagnose the hidden fault characteristics of rotating machinery than the general method. In this paper, the holographic spectrum analysis technology of rotating machinery is taken as the research object, the analysis accuracy of the holographic spectrum under dense spectrum and noise is studied and improved, and the harmonics wavelet axis track is tracked and purified. Finally, the holographic spectrum analysis system is developed. The research work is as follows: 1 on the basis of the detailed introduction of the holographic spectrum testing principle, the rotational speed curve is calculated based on the first order digital differential and least square smoothing method, which provides the guarantee for the accurate acquisition of power frequency. At the same time, by analyzing the principle of ratio correction method and phase difference correction method with stronger anti-noise ability, it is found that two kinds of frequency domain interpolation algorithms can not achieve intensive spectrum correction. The necessity of eliminating or weakening the influence of dense spectrum and noise on the accuracy of holographic spectrum analysis is emphasized. 2 in order to reduce the precision of holographic spectrum in dense spectrum, the holographic spectrum based on spectrum refinement and phase difference correction is studied and realized. It uses the method of complex analytic bandpass filter to refine the local interval centered on the order frequency, and eliminates the influence of the dense frequency spectrum by improving the frequency resolution. The phase difference method is used to correct the frequency, amplitude and phase of the maximum spectral peak in the thinning region. The amplitude and phase information of each order are extracted accurately. Finally, the holographic spectrum is synthesized. Simulation and field experiments show that the holographic spectrum analysis method based on spectrum refinement and phase difference correction is correct and effective. (3) tracking the research status of axis trajectory purification at home and abroad, on the basis of synthetically analyzing the advantages and disadvantages of the existing methods, the harmonic wavelet is selected to purify the holographic spectrum analysis system, and the algorithm process is realized. A simulation example is given to verify the correctness and effectiveness of the algorithm. (4) the holographic spectrum analysis system of rotating machinery is developed by using object-oriented programming technology and VC development platform, and the functional modules of the system are tested with experimental data. The test results show that the holographic spectrum analysis system developed in this paper has good engineering application value.
【学位授予单位】：重庆大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：TH165.3
【引证文献】
相关硕士学位论文 前2条
1 杨才源;离散全矢谱校正方法及工程应用研究[D];郑州大学;2013年
2 王绍;全矢高阶谱分析关键技术研究[D];郑州大学;2013年
，本文编号：2413744
本文链接：https://www.wllwen.com/kejilunwen/jixiegongcheng/2413744.html
