由于具有较小的旁瓣，汉宁、汉明和三角窗常被用于振动信号频谱分析中。但是，加这些窗会导致落在窗边缘的脉冲信号受到抑制而无法被有效检测出来，而脉冲信号中可能含有设备运行重要信息。针对此问题，研究了基于多重重叠技术的振动信号频谱分析方法并提出了相对最大值抽取的多重重叠算法。通过虚拟仿真信号和MFS-MG机械故障仿真试验台实测信号对多重重叠频谱分析方法进行了试验测试。虚拟仿真结果表明，边缘脉冲信号的多重重叠频谱幅值为一般频谱分析方法的8-10倍；试验台故障模拟信号的分析结果同样表明，多重重叠频谱分析方法比一般频谱分析方法更能表现边缘脉冲信号的宽频特征。
Abstract
A relative maximum extraction algorithm was submitted and combined with overlap processing method to detect the pulse signal in the edge of a time series. The pulse signal was often suppressed in case of hanning, hamming and triangular windows, with a low side-lobe, used in spectrum analysis of vibration signal processing. But the pulse signal always contained some important information of a running machine. Signals produced by function generator and machine fault simulator(MFS-MG) were used in the above signal processing method. It shows that the overlap processing spectrum result of virtual simulation test is about 8 to 10 times to the result from traditional spectrum analysis. And the overlap processing spectrum of pulse signal which produced by MFS-MG can be unfold clearly, especially when it’s compared with traditional spectrum analysis.
关键词
多重重叠技术 /
相对最大值抽取 /
频谱分析 /
振动测试
{{custom_keyword}} /
Key words
overlap processing /
relative maximum extraction /
spectrum analysis /
vibration test
{{custom_keyword}} /
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] Lotfi Saidi, Jaouher Ben Ali, Farhat Fnaiech. Application of higher order spectral features and support vector machines for bearing fault sclassification[J]. ISA Transactions,2015,54(1):193-206
[2] Boštjan Dolenc, Pavle Boškoski, Đani Juričić. Distributed bearing fault diagnosis based on vibration analysis[J]. Mechanical Systems and Signal Processing,2016,66(67): 521-532
[3] H Saruhan, S Sandemir, A Çiçek, et al. Vibration Analysis of Rolling Element Bearings Defects[J]. Journal of Applied Research and Technology,2014,12(3):384-395
[4] 王艳丰,朱靖,滕光蓉等.航空发动机转子早期裂纹故障振动特征的 维谱分析[J].振动与冲击,2015,34(1),88-93
WANG Yan-feng, ZHU Jing, TENG Guang-rong, et al. －dimension spectrum analysis on early cracked fault characters of aero engine rotors[J]. Journal of Vibration and Shock,2015,34(1),88-93
[5] M W TRETHEWEY. WINDOW AND OVERLAP PROCESSING EFFECTS ON POWER ESTIMATES FROM SPECTRA[J]. Mechanical Systems and Signal Processing, 2000,14(2):267-278
[6] 明建成,祁舒喆.实时频谱分析仪中重叠帧频谱分析技术研究[J].电子测量技术(理论与算法),2013,36(12),32-34
Ming Jiancheng, Qi Shuzhe. Study of overlapping frames spectrum analysis techniques in real-time spectrum analyzer[J]. Electronic Measurement Technology, 2013,36(12),32-34
[7] 杨凯,邹耘,尹路明.重叠加窗FFT频域抗干扰算法设计[J].测绘科学技术学报,2013,30(3),228-231
YANG Kai, ZOU Yun, YIN Luming.Design of Frequency Domain Anti-Jamming AlgorithmBased on Overlap Windowed FFT[J]. Journal of Geomatics Science and Technology,2013,30(3),228-231
[8] Brincker. Rune, Brandt. Anders, FFT integration of time series using an overlap-add technique[C]. 28th IMAC, A Conference on Structural Dynamics, Jacksonville, FL, United states, 2010 : 1467-1474
[9] Wang Sichun, Inkol Robert, Rajan Sreeraman, Patenaude Francois. A comparison of the normalized detection threshold for the overlapped and non-overlapped FFT summation detectors[C]. 2011 Canadian Conference on Electrical and Computer Engineering, Niagara Falls, ON, Canada, 2011 : 000136-000141
[10] Uchida Shoya, Sato Akihiro, Inamori Mamiko, et al. Overlapped FFT scheme with frames of non-continuous samples[C]. 2012 IEEE International Conference on Communication Systems, ICCS 2012, 2012:90-94
{{custom_fnGroup.title_cn}}
脚注
{{custom_fn.content}}
PDF(1478 KB)
