A novel scheme utilizing fractional time-frequency local binary pattern(LBP) spectrum was proposed to express time-frequency characteristics of gear fault signal. Since the time-frequency resolution of S transform was low for high-frequency signals, fractional Fourier transform with excellent time-frequency rotating character was integrated into S transform and as a result a fractional S transform was designed. Fractional S transform was employed to obtain 2D time-frequency representations of gear fault signals. Then, LBP operator was introduced and imposed on time-frequency images from fractional S transform to extract fractional time-frequency LBP spectrum. With the concept of uniform pattern LBP and separability evaluation criterion, the fractional time-frequency LBP spectrum was optimally selected for succinct description of gear fault features. Gear vibration signals from five different states were studied. The results indicate that optimally selected fractional time-frequency LBP spectrum is of outstanding feature description performance. It is a kind of novel and effective feature parameters for gear fault signals.
[1] 李兵, 米双山, 刘鹏远, 等. 二维非负矩阵分解在齿轮故障诊断中的应用[J]. 振动、测试与诊断, 2012, 32(5): 836-840.
LI Bing, MI Shuang-shan, LIU Peng-yuan, et al. Application of two-dimensional non-negative factorization for gear fault diagnosis[J]. Journal of Vibration, Measurement & Diagnosis, 2012, 32(5): 836-840.
[2] 姜鸣, 陈进, 汪慰军. 几种Cohen 类时频分布的比较及应用 [J]. 机械工程学报, 2003, 39(8): 129-134.
JIANG Ming, CHEN Jin, WANG Wei-jun. Comparison and application of some time-frequency distributions belonging to Cohen class[J]. Chinese Journal of Mechanical Engineering, 2003, 39(8): 129-134.
[3] Meltzer G, Ivanov Y Y. Fault detection in gear drives with non-stationary rotational speed——part I: the time-frequency approach[J]. Mechanical Systems and Signal Processing, 2003, 17(5): 1033-1047.
[4] 马伦, 康建设, 孟妍, 等. 基于Morlet小波变换的滚动轴承早期故障特征提取研究[J]. 仪器仪表学报, 2013, 34(4): 920-926.
MA Lun, KANG Jian-she, MENG Yan, et al. Research on feature extraction of rolling bearing incipient fault based on Morlet wavelet transform[J]. Chinese Journal of Scientific Instrument, 2013, 34(4): 920-926.
[5] 赵淑红, 朱光明. S变换时频滤波去噪方法[J]. 石油地球物理勘探, 2007, 42(4): 402-408.
ZHAO Shu-hong, ZHU Guang-ming. Time-frequency filtering to denoise by S transform[J]. Oil Geophysical Prospecting, 2007, 42(4): 402-408.
[6] 邵忍平, 曹精明, 李永龙. 基于EMD 小波阈值去噪和时频分析的齿轮故障模式识别与诊断[J]. 振动与冲击, 2012, 31(8): 96-101.
SHAO Ren-ping, CAO Jing-ming, LI Yong-long. Gear fault pattern identification and diagnosis using time-frequency analysis and wavelet threshold de-noising based on EMD. [J]. Journal of Vibration and Shock, 2012, 31(8): 96-101.
[7] 陶然, 邓兵, 王越. 分数阶傅里叶变换及其应用[M]. 北京:清华大学出版社, 2009.
TAO Ran, DENG Bing, WANG Yue. Fractional Fourier Transform and Its Applications[M]. Beijing: Tsinghua University Press, 2009.
[8] Ervin S, Igor D, Ljubisa S. Fractional Fourier transform as a signal processing tool: An overview of recent development[J]. Signal Processing, 2011, 91: 1351-1369.
[9] Loris N, Alessandra L, Sheryl B. Survey on LBP based texture descriptors for image classification[J]. Expert Systems with Applications, 2012, 39: 3634-3641.
[10] SONG Tie-cheng, LI Hong-liang. WaveLBP based hierarchical features for image classification[J]. Pattern Recognition Letters 2013, 34: 1323-1328.
[11] Ojala T, Pietikainen M, Maeenpaa T. Multiresolution gray-scale and rotation invariant texture classification with local binary patterns[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(7): 971-987.
[12] 周晓英, 李巍华, 丁康.基于关联向量机的齿轮故障检测方法研究[J]. 振动与冲击, 2008, 27(6): 51-54.
ZHOU Xiao-ying, LI Wei-hua, DING Kang. Gear fault detection based on relevance vector machine[J]. Journal of Vibration and Shock,, 2008, 27(6): 51-54.
[13] Djurovi I, Sejdi E, Jiang J. Frequency-based window width optimization for S-transform[J]. AEU-International Journal of Electronics and Communications, 2008, 62(4): 245–250.
[14] 白航, 赵拥军, 胡德秀. 时频图像局部二值模式特征在雷达信号分类识别中的应用[J]. 宇航学报, 2013, 34(1): 139-146.
BAI Hang, ZHAO Yong-jun, HU De-xiu. Radar signal recognition based on the local binary pattern feature of time-frequency image[J]. Journal of Astronautics, 2013, 34(1): 139-146.