基于高阶累积量图像特征的柴油机故障诊断研究

沈虹1, 2,赵红东2,梅检民1,曾锐利1

振动与冲击 ›› 2015, Vol. 34 ›› Issue (11) : 133-138.

PDF(1781 KB)
PDF(1781 KB)
振动与冲击 ›› 2015, Vol. 34 ›› Issue (11) : 133-138.
论文

基于高阶累积量图像特征的柴油机故障诊断研究

  • 沈虹1, 2,赵红东2,梅检民1,曾锐利1
作者信息 +

Research on Diesel Engine Fault Diagnosis Based on High-order Cumulant Image Feature

  • SHEN Hong1,2,ZHAO Hong-dong2,MEI Jian-min1,ZENG Rui-li1
Author information +
文章历史 +

摘要

针对发动机不同部位的机械故障特征容易混淆,且往往淹没在其他分量和强噪声中难于区分和提取的问题,提出了一种基于高阶累积量图像特征的柴油机故障诊断方法。截取柴油机6个工作循环的振动信号分别进行三阶累积量计算,累加平均得到1个工作循环信号的三阶累积量,提取柴油机不同故障状态基于三阶累积量图像灰度共生矩阵的图像纹理特征参数,利用支持向量机进行模式识别。试验结果表明:该方法既能抑制噪声干扰,又能充分利用高阶累积量图像中的纹理特征信息分析非稳态信号,提取的特征参数能有效识别发动机6种技术状态,与传统的基于高阶累积量的特征提取相比,提高了故障诊断准确率。

Abstract

Against the problem of different mechanical fault feature of the diesel engine is easy to confused and is often drowned in other components and color noises, so it is difficult to distinguish and extract the fault feature, a fault diagnose method based on high-order cumulant image feature is proposed. The three-order cumulant of six circulation of vibration signal is calculated respectively, and the result is averaged to get three-order cumulant of one circulation, the image texture features parameter based on three-order cumulant image gray level co-occurrence matrix (GLCM) of different diesel engine status is extracted, the pattern recognition result is obtained by support vector machine (SVM). Experimental result shows that this method can inhibit noises and make full use of texture feature information of high-order cumulant image to analysis unstable signals, the extracted feature can distinguish 6 kinds status of diesel engine effectively, the fault diagnosis accuracy is increased compared with the traditional feature extraction based on high-order cumulant.

关键词

三阶累积量 / 图像纹理特征 / 灰度共生矩阵 / 特征提取

Key words

three-order cumulant / image texture feature / gray level co-occurrence matrix / feature extraction

引用本文

导出引用
沈虹1, 2,赵红东2,梅检民1,曾锐利1. 基于高阶累积量图像特征的柴油机故障诊断研究[J]. 振动与冲击, 2015, 34(11): 133-138
SHEN Hong1,2,ZHAO Hong-dong2,MEI Jian-min1,ZENG Rui-li1. Research on Diesel Engine Fault Diagnosis Based on High-order Cumulant Image Feature[J]. Journal of Vibration and Shock, 2015, 34(11): 133-138

参考文献

[1] 肖云魁.汽车故障诊断学[M].北京:北京理工大学出版社,2006.
[2]. 沈虹,赵红东,梅检民,等.基于角域四阶累积量切片谱的柴油机连杆轴承故障特征提取[J] .振动与冲击,2014,33(11) :90-94.
SHEN Hong, ZHAO Hong-dong, MEI Jian-min, et al. Fault feature extraction of connecting rod bearings of a diesel engine based on angle domain fourth-order cumulant slice spectrum[J]. Journal of Vibration and Shock, 2014, 33(11):90-94. 
[3] 张玲玲,梅检民,贾继德,等.柴油机加速振动信号的阶比双谱特征提取[J].振动与冲击,2013,32(1):154-158.
ZHANG Ling-ling, MEI Jian-min, JIA Ji-de. et al. Feature extraction for accelerating vibration signal of a diesel engine based on order bi-spectrum estimation[J]. Journal of Vibration and Shock, 2013, 32(1):154-158.
[4] 夏天,王新晴,赵慧敏,等.基于高阶累积量的柴油发动机曲轴轴承故障特征提取[J].振动与冲击,2011,30(1):77-80.
XIA Tian, WANG Xin-qing, ZHAO Hui-min, et al. Extracting fault features of a Diesel engine’s crankshaft bearing based on high-order cumulation[J]. Journal of Vibration and Shock, 2011, 30(1):77-80.
[5] 孙云岭,朴甲哲,张永祥. 基于瞬时转速和双谱的内燃机故障诊断研究[J]. 内燃机学报,2004,22(3):241-244.
SUN Yun-ling, PIAO Jia-zhe, ZHANG Yong-xiang. Research on fault diagnosis for internal combustion engines using transient rotate speed and bispectrum [J]. Transactions of CSICE. 2004, 22(3):241-244.
[6] 梅检民,张玲玲,肖云魁,等.基于高阶累积量的轴承并发故障振动信号分析[J].内燃机学报,2011,29(4):327-331.
MEI Jian-min, ZHANG Ling-ling, XIAO Yun-kui. et al. Analysis of vibration signals of bearing concurrent fault based on higher order cumulants[J]. Transactions of CSICE, 2011, 29(4):327-331.
[7] 李智,陈祥初,刘政波.基于图像处理与神经网络的内燃机故障诊断研究[J].内燃机工程,2011,19(4):356-360.
LI Zhi, CHEN Xiang-chu, LIU Zheng-bo. Study of diagnosis methods for diesel’s faults based on picture-processing and neural network[J]. Chinese Internal Combustion Engine Engineering, 2011, 19(3):241-244.
[8] 王成栋,张优云,夏勇.模糊函数图像在柴油机气阀故障诊断中的应用研究[J].内燃机学报,2004,22(2):162-168.
WANG Cheng-dong, ZHANG You-yun, XIA Yong. Study on the use of ambiguity function images in fault diagnosis for diesel valve train[J]. Transactions of CSICE, 2004, 22(2):162-168.
[9] 蔡奇志,黄宜坚.自回归三谱切片研究[J].仪器仪表学报,2009,30(2):345-350.
CAI Qi-zhi, Huang Yi-jian. Study on autoregressive trispectrum slices[J]. Chinese Joumal of Scientific Instrument, 2009, 30(2): 345-350.
[10] BARAEDI A, PARMIGGIAN F. An investigation of texture characteristics associated with gray level co-occurrence matrix statistical parameters[J]. IEEE Trans. On geo-science and remote sensing, 1995, 33(2):293-303.

PDF(1781 KB)

Accesses

Citation

Detail

段落导航
相关文章

/