GAO Yuan， LI Yi-bo
（National Key Laboratory of Precision Testing Techniques and Instrument, Tianjin University, Tianjin 300072, China）
Abstract： Due to non-stationary characteristics of the vibration signal acquired from cylinder head, a misfire fault diagnosis system of automobile engines based on correlation coefficient gained by wavelet packet and extreme learning machine (ELM) is proposed. Firstly, the original signal is decomposed by wavelet packet, and correlation coefficients between the reconstructed signal of each sub-band and the original signal as well as the energy entropy of each sample are obtained. Then, the eigenvectors established by the correlation coefficients method and the energy entropy method fused with kurtosis are inputted to the four kinds of classifiers including BP neural network, KNN classifier, support vector machine and ELM respectively for training and testing. Experimental results show that the method proposed in this paper can effectively reflect the differences that the fault produces and identify the single-cylinder misfire accurately, which has the advantages of higher accuracy and shorter training time.
Key words： automobile engine; wavelet packet; correlation coefficient; extreme learning machine (ELM); misfire fault identification
CLD number： TH17 Document code： A
Article ID： 1674-8042（2017）04-0384-12 doi： 10.3969/j.issn.1674-8042-2017-04-012
References
［1］ZHANG Zhen-dong, WANG Bo-nian, HAN Bai-shun, et al. Misfire detection and evaluation method for automobile engine. Internal Combustion Engines, 1999, (6): 12-15.
［2］XU Xiao-jie, WANG Jing, CAI Wen-yuan. Research on engine misfire. Microcomputer Information, 2008, 24(11): 174-176.
［3］YU He-ji. Engineering application of vibration diagnosis. Beijing: Metallurgical Industry Press, 2000.
［4］Devasenapati S B, Sugumaran V, Ramachandran K I. Mi-sfire identification in a four-stroke four-cylinder petrol engine using decision tree. Expert Systems with Applications, 2010, 37(3): 2150-2160.
［5］HU Jie, YAN Fu-wu. A research on the misfire diagnosis method of gasoline engine based on BP neural network. Automotive Engineering, 2011, 33(2): 101-105.
［6］WANG Yu, CHU Jiang-wei. Diagnosis information extraction of misfire fault of vehicle electronically controlled engine based on Daubechies wavelet. Forest Engineering, 2014, 30(2): 138-142.
［7］WANG Zi-jian, WANG De-jun. Engine misfire diagnosis based on probabilistic neural network. Journal of Jilin University, 2016, 34(2): 229-236.
［8］Freescale R&D Center. Wavelet analysis and MATLAB7 implementation. Beijing: Publishing House of Electronics Industry, 2005.
［9］SI Jing-ping, NIU Jia-hua, GUO Li-na, et al. Application of EEMD and SVM in engine fault diagnosis. Vehicle Engine, 2015, (1): 81-86.
［10］Huang G B, Zhu Q Y, Siew C K. Extreme learning machine:Theory and applications. Neurocomputing, 2006, 70(1): 489-501.
［11］Huang G B, Zhu Q Y, Siew C K. Extreme learning machine:A new learning scheme of feedforward neural networks. In: Proceedings of 2004 IEEE International Joint Conference on Neural Networks. Budapest, Hungary, 2004: 985-990.
［12］YAN Yu. Fault diagnosis and the selection of wavelet base about electromotor. Taiyuan: Taiyuan University of Technology, 2006.
［13］DU Can-yi, DING Kang, YANG Zhi-jian. Feature extraction of engine misfire fault based on finite element and multi-body dynamics simulation. Journal of Vibration and Shock, 2012, 31(9): 18-23.
基于小波包和极限学习机的汽车发动机失火故障识别
高远， 李一博
（天津大学 精密测量技术与仪器国家重点实验室， 天津 300072）
摘要:针对缸盖振动信号的非平稳特性， 提出了基于小波包相关系数和极限学习机的汽车发动机失火故障诊断系统。 首先， 对原始信号进行小波包分解， 然后计算得到每个样本的能量熵和每个样本各子频带重构信号与原始信号的相关性系数。 分别利用相关系数法和能量熵融合峭度的方法建立特征向量， 随后输入到BP神经网络和极限学习机中进行训练和测试。 实验结果表明， 该方法可以有效地反映故障产生的差异并准确地识别单缸失火故障， 具有精度高、 训练时间短的优点。
关键词:汽车发动机； 小波包； 相关系数； 极限学习机； 失火故障识别
引用格式：GAO Yuan， LI Yi-bo. Misfire identification of automobile engines based on wavelet packet and extreme learning machine. Journal of Measurement Science and Instrumentation, 2017, 8(4)： 384-395. ［doi： 10.3969／j.issn.1674-8042.2017-04-012］
[full text view]
