基于全矢谱的齿轮系统故障诊断方法研究
发布时间：2018-11-03 14:04
【摘要】：随着科学技术的发展与现代工业的进步,各种机械设备日趋大型化、高速化和重载化,自动化程度也越来越高。齿轮因其承载能力大、传动精度高、恒功率传动等特点,被广泛地应用于各种旋转机械。齿轮的运行好坏直接决定了机械设备的安全与稳定。因此,对齿轮系统的故障机理及诊断方法研究具有重要意义。 由于转子的涡动特性,使得同一截面相互垂直的两个通道数据可能存在偏差。若只采用单通道的数据进行诊断,获得的故障特征信息并不完整,影响故障诊断的准确性与可靠性。全信息技术有效的融合了两个通道的信息,融合后的信息无论从结构上还是能量上都能够真实的反映转子的实际状态。 本文重点介绍了全信息技术中的全矢谱技术。全矢谱技术处理方法具有高分辨率的特点,并能进一步做能量与三维分析,还可以拓展出一系列的分析方法,如矢功率谱、矢倒谱、短时矢谱等。本文主要以全矢谱技术为基础,结合传统分析方法,以拓展全矢谱分析方法,并进行齿轮故障诊断。 当齿轮发生故障时,采集到的齿轮振动信号往往是非平稳信号。矢双谱可有效的分析非平稳信号,提取齿轮的故障特征。但矢双谱只能抑制高斯噪声,却对非高斯噪声却无能为力。将小波包变换与矢双谱结合,提出小波包—矢双谱。利用小波包变换的降噪功能预处理原始信号,然后再进行矢双谱分析。研究表明,该方法能有效的消除噪声。 抑制了交叉项后的Wigner高阶谱可从各个角度对非平稳信号进行分析,真实的再现了信号频率成分随时间的演变过程,为齿轮故障特征提取及故障诊断提供有力的工具。将全矢谱技术与Wigner高阶谱结合,提出矢Wigner高阶谱,并将其应用到齿轮故障诊断中。仿真及实例表明,该方法不仅发挥了Wigner高阶谱的优点,而且融合了同源双通道信息,能够全面的反映齿轮的振动特征。
[Abstract]:With the development of science and technology and the progress of modern industry, all kinds of machinery and equipment are becoming more and more large, high speed and heavy load, and the degree of automation is becoming higher and higher. Gear is widely used in various rotating machinery because of its high bearing capacity, high transmission precision, constant power transmission and so on. Gear operation directly determines the safety and stability of mechanical equipment. Therefore, it is of great significance to study the fault mechanism and diagnosis method of gear system. Due to the vortex characteristics of the rotor, the data of two channels perpendicular to each other in the same section may be deviated. If only single channel data is used for fault diagnosis, the obtained fault feature information is incomplete, which affects the accuracy and reliability of fault diagnosis. The full information technology effectively integrates the information of two channels, and the fused information can reflect the actual state of the rotor in both structure and energy. This paper focuses on the full vector spectrum technology in the full information technology. The full vector spectrum processing method has the characteristics of high resolution and can be used for further energy and 3D analysis. A series of analytical methods such as vector power spectrum vector cepstrum and short time vector spectrum can also be developed. This paper mainly based on the whole vector spectrum technology, combined with the traditional analysis method, to expand the total vector spectrum analysis method, and to carry out gear fault diagnosis. When gear failure occurs, the vibration signal of gear is usually non-stationary signal. Vector bispectrum can effectively analyze non-stationary signals and extract fault features of gears. But sagittal bispectrum can only suppress Gao Si noise, but it can not do anything to non-Gao Si noise. By combining wavelet packet transform with vector bispectrum, a wavelet packet vector bispectrum is proposed. The denoising function of wavelet packet transform is used to preprocess the original signal, and then the vector bispectral analysis is carried out. The results show that this method can eliminate noise effectively. The Wigner higher order spectrum after suppressing the cross term can analyze the non-stationary signal from various angles, and reproduce the evolution process of the signal frequency component with time, which provides a powerful tool for gear fault feature extraction and fault diagnosis. A vector Wigner high order spectrum is proposed by combining full vector spectrum with Wigner high order spectrum, and it is applied to gear fault diagnosis. Simulation and an example show that this method not only takes advantage of Wigner high order spectrum, but also combines the information of homologous two channels and can fully reflect the vibration characteristics of gears.
【学位授予单位】：郑州大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：TH132.41;TH165.3
本文编号：2308008
[Abstract]:With the development of science and technology and the progress of modern industry, all kinds of machinery and equipment are becoming more and more large, high speed and heavy load, and the degree of automation is becoming higher and higher. Gear is widely used in various rotating machinery because of its high bearing capacity, high transmission precision, constant power transmission and so on. Gear operation directly determines the safety and stability of mechanical equipment. Therefore, it is of great significance to study the fault mechanism and diagnosis method of gear system. Due to the vortex characteristics of the rotor, the data of two channels perpendicular to each other in the same section may be deviated. If only single channel data is used for fault diagnosis, the obtained fault feature information is incomplete, which affects the accuracy and reliability of fault diagnosis. The full information technology effectively integrates the information of two channels, and the fused information can reflect the actual state of the rotor in both structure and energy. This paper focuses on the full vector spectrum technology in the full information technology. The full vector spectrum processing method has the characteristics of high resolution and can be used for further energy and 3D analysis. A series of analytical methods such as vector power spectrum vector cepstrum and short time vector spectrum can also be developed. This paper mainly based on the whole vector spectrum technology, combined with the traditional analysis method, to expand the total vector spectrum analysis method, and to carry out gear fault diagnosis. When gear failure occurs, the vibration signal of gear is usually non-stationary signal. Vector bispectrum can effectively analyze non-stationary signals and extract fault features of gears. But sagittal bispectrum can only suppress Gao Si noise, but it can not do anything to non-Gao Si noise. By combining wavelet packet transform with vector bispectrum, a wavelet packet vector bispectrum is proposed. The denoising function of wavelet packet transform is used to preprocess the original signal, and then the vector bispectral analysis is carried out. The results show that this method can eliminate noise effectively. The Wigner higher order spectrum after suppressing the cross term can analyze the non-stationary signal from various angles, and reproduce the evolution process of the signal frequency component with time, which provides a powerful tool for gear fault feature extraction and fault diagnosis. A vector Wigner high order spectrum is proposed by combining full vector spectrum with Wigner high order spectrum, and it is applied to gear fault diagnosis. Simulation and an example show that this method not only takes advantage of Wigner high order spectrum, but also combines the information of homologous two channels and can fully reflect the vibration characteristics of gears.
【学位授予单位】：郑州大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：TH132.41;TH165.3
【引证文献】
相关硕士学位论文 前3条
1 尚慧娟;面向全矢谱分析的转子动态故障特性研究[D];郑州大学;2012年
2 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
3 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：2308008
本文链接：https://www.wllwen.com/kejilunwen/jixiegongcheng/2308008.html
