基于集合经验模态分解-支持向量机的高压共轨系统故障诊断方法

doi:10.12382/bgxb.2021.0155

摘要/Abstract

摘要： 柴油机高压共轨系统运行时轨压波动信号波动较大且非线性特征较为明显，使其故障诊断较为困难。针对高压共轨系统轨压信号状态参数难以提取与识别的问题，提出一种基于集合经验模态分解（EEMD）—支持向量机（SVM）的故障诊断方法。通过EEMD将轨压信号分解为一系列固有模态函数，利用过零率曲线确定的特征提取准则提取本征模态函数中的特征值。将提取的特征值输入SVM中进行故障类型的诊断。通过AME Sim软件仿真实验获得轨压信号，对比7种不同的特征值选择方法，最终选取能量特征值构建特征值向量并进行识别和诊断结果分析，以验证该方法的正确性与准确性。结果表明：所提出的基于EEMD—SVM的高压共轨系统故障诊断方法能够对6种不同的运行状态进行状态识别，平均故障诊断正确率可达96.11%。

关键词: 高压共轨系统, 故障诊断, 集合经验模态分解, 支持向量机

Abstract: When the high-pressure common rail system for diesel engine is running， the rail pressure fluctuation signal fluctuates greatly and has obvious nonlinear characteristics， which makes the fault diagnosis more difficult. For the problem that the state parameters of rail pressure signal in high-pressure common rail system are difficult to extract and identify， a fault diagnosis method based on ensemble empirical mode decomposition (EEMD)-support vector machine (SVM) is proposed. The rail pressure signal is decomposed into a series of eigenmode functions by EEMD， and the eigenvalues in the eigenmode functions are extracted using the feature extraction criterion determined by the zero-crossing rate curve. The extracted eigenvalues are input into SVM for fault type diagnosis. The rail pressure signal is obtained through AMESim software simulation experiment， and seven different eigenvalue selection methods are compared. Finally， the energy eigenvalue is selected to construct the eigenvalue vector for identification， and the diagnosis results are analyzed to verify the correctness and accuracy of the proposed method. The results show that the proposed EEMD-SVM-based fault diagnosis method for high-pressure common rail system can be used to identify six different operating states， with the average fault diagnostic accuracy rate of 96.11%.

Key words: high-pressurecommonrailsystem, faultdiagnosis, ensembleempiricalmodedecomposition, supportvectormachine

中图分类号:

TK428

李良钰，苏铁熊，马富康，吴小军，徐春龙. 基于集合经验模态分解-支持向量机的高压共轨系统故障诊断方法[J]. 兵工学报, 2022, 43(5): 992-1001.

LI Liangyu， SU Tiexiong， MA Fukang， WU Xiaojun， XU Chunlong. Fault Diagnosis Method of High-pressure Common Rail System Based on EEMD-SVM[J]. Acta Armamentarii, 2022, 43(5): 992-1001.

参考文献

［1］樊高瞻，周俊，朱昆莉.基于改进形态-小波阈值降噪的轴承复合故障声学诊断［J］.振动与冲击，2020，39(12):221-226，288.
FAN G Z，ZHOU J，ZHU K L.An improved morphological-wavelet threshold de-noising method based acoustic diagnosis for bearing composite faults［J］.Journal of Vibration and Shock，2020，39(12): 221-226，288.（in Chinese）
［2］张健，张子阳，起雪梅，等.基于EMD和VPMCD的滚动轴承故障诊断［J］.机械设计，2019，36(增刊2):91-94.
ZHANG J，ZHANG Z Y，QI X M，et al. Fault diagnosis of rolling bearing based on EMD and VPMCD［J］.Journal of Mechane Design，2019，36(S2):91-94.(in Chinese)
［3］赵荣珍，张琛，邓林峰.基于EEMD与模糊信息熵的旋转机械故障诊断方法［J］.兰州理工大学学报，2020，46(3):39-44.
ZHAO R Z，ZHANG C，DENG L F. Fault diagnosis method of rotating machinery based on EEMD and fuzzy information entropy［J］.Journal of Lanzhou University of Technology，2020，46(3):39-44. (in Chinese)
［4］柴艳有.基于核学习理论的船舶柴油机故障诊断研究［D］. 哈尔滨：哈尔滨工程大学，2012.
CHAI Y Y.Research on fault diagnosis of marine diesel engine based on kernel-based learning theory［D］.Harbin: Harbin Engineering University，2012.(in Chinese)
［5］蒋佳炜，胡以怀，柯赟，等.基于小波包特征提取和模糊熵特征选择的柴油机故障分析［J］.振动与冲击，2020，39(4):273-277，298.
JIANG J W，HU Y H，KE Y，et al.Fault analysis of diesel engines based on wavelet packet energy spectrum feature extraction and fuzzy entropy feature selection［J］. Journal of Vibration and Shock，2020，39(4):273-277，298.(in Chinese)
［6］孟庆宇.基于EEMD-SVM的自动扶梯主驱动轴轴承故障诊断方法研究［J］.机械与电子，2020，38(5):51-53，58.
MENG Q Y. Research on fault diagnosis method for main drive shaft bearing of escalator based on EEMD-SVM［J］. Machinery and Electronics，2020，38(5):51-53，58.(in Chinese)
［7］郑小霞，周国旺，任浩翰，等.基于变分模态分解和排列熵的滚动轴承故障诊断［J］.振动与冲击，2017，36(22):22-28.
ZHENG X X，ZHOU G W，REN H H，et al.A rolling bearing fault diagnosis method based on variational mode decomposition and permutation entropy［J］.Journal of Vibration and Shock，2017，36(22): 22-28.(in Chinese)
［8］郭伟超，赵怀山，李成，等.基于小波包能量谱与主成分分析的轴承故障特征增强诊断方法［J］.兵工学报，2019，40(11):2370-2377.
GUO W C，ZHAO H S，LI C，et al.Fault feature enhancement method for rolling bearing fault diagnosis based on wavelet packet energy spectrum and principal component analysis［J］.Acta Armamentarii，2019，40(11): 2370-2377.(in Chinese)
［9］李志农，胡志峰，毛清华，等.非线性调频模态分解-同步提取变换方法及其在滚动轴承故障诊断中的应用［J］.兵工学报，2021，42(6):1324-1330.
LI Z N，HU Z F，MAO Q H，et al.Variational nonlinear chirp mode decomposition-synchroextracting transform method and its application in fault diagnosis of rolling bearings［J］.Acta Armamentarii，2021，42(6):1324-1330.(in Chinese)
［10］李志农，朱明.基于变分模态分解的机械故障诊断方法研究［J］.兵工学报，2017，38(3):593-599.
LI Z N，ZHU M.Research on mechanical fault diagnosis method based on variational mode decomposition［J］.Acta Armamentarii，2017，38(3): 593-599.(in Chinese)
［11］ BOSER B E，GUYON I M，VAPNIK V N.A training algorithm for optimal margin classifiers［C］∥Proceedings of the 5th Annual Workshop on Computational Learning Theory. Pittsburgh，PA，US：ACM，1992：144-152.
［12］ HOLLANDJ H. Adaptation in natural and artificial systems: an introductory analysis with applications to biology，control，and artificial intelligence［M］.Cambridge，MA，US:MIT Press，1992.
［13］李海涛，邓少贵，王跃祥，等.基于经验模态分解的核磁共振去噪方法研究［J］.西南石油大学学报(自然科学版)，2020，42(3):51-59.
LI H T，DENG S G，WANG Y X，et al. Research on NMR denoising method based on empirical mode decomposition［J］.Journal of Southwest Petroleum University (Science & Technology Edition)，2020，42(3):51-59.(in Chinese)
［14］李卿.EMD时频分析方法的研究与应用［D］.武汉：华中师范大学，2008.
LI Q.Research and application of EMD time-frequency analysis method［D］.Wuhan:Central China Normal University，2008.(in Chinese)
［15］ ZHANG J，TIAN J，WEN T，et al.Deep fault diagnosis for rotating machinery with scarce labeled samples［J］. Chinese Journal of Electronics，2020，29(4):693-704.
［16］ FERNNDEZ-DELGADO M，CERNADAS E，BARRO S，et al.Do we need hundreds of classifiers to solve real world classification problems［J］. The Journal of Machine Learning Research，2014，15(1):3133-3181.
［17］ CARUANA R，KARAMPATZIAKIS N， YESSENALINA A. An empirical evaluation of supervised learning in high dimensions［C］∥Proceedings of the 25th International Conference on Machine Learning.New York，NY，US: Association for Computing Machinery，2008.
［18］ WANG G，WANG T Y，ZHENG H T，et al.Man vs. machine: practical adversarial detection of malicious crowdsourcing workers［C］∥Proceedings of the 23rd USENIX Security Symposium.San Diego，CA，US: USENIX Association，2014：239-254.
［19］孔安生，王洪澄，李国正.神经网络集成与支持向量机在多值分类问题上的比较研究［J］.计算机工程与应用，2005，41(1): 46-47，119.
KONG A S，WANG H C，LI G Z. Comparison study on neural network ensemble and support vector machines for multiple classification problems［J］.Computer Engineering and Applications，2005，41(1):46-47，119.(in Chinese)
［20］武哲，张强，黄华蒙，等.基于多尺度排列熵的复合行星齿轮故障诊断研究［J］.机械设计与制造，2020(9):182-186.
WU Z，ZHANG Q，HUANG H M，et al. Research on fault diagnosis of compound planetary gears based on dynamic model and multiscale permutation entropy［J］. Mechanical Design and Manufacture，2020(9):182-186.(in Chinese)
［21］ LEI Y G，LIN J，HE Z J，et al. A review on empirical mode decomposition in fault diagnosis of rotating machinery［J］.Mechanical Systems and Signal Processing，2013，35(1/2):108-126.
［22］李如强，陈进，伍星.基于奇异值分解、模糊聚类和粗糙集理论的旋转机械故障诊断［J］.振动与冲击，2005，24(4):46-49.
LI R Q，CHEN J，WU X.Fault diagnosis of rotating machinery based on SVD，FCM and RST［J］.Journal of Vibration and Shock，2005，24(4):46-49. (in Chinese)
［23］林煜凯，张会林.基于EEMD和ICS-SVM的齿轮箱故障诊断研究［J］.电力科学与工程，2020，36(10):56-62.
LIN Y K，ZHANG H L. Study on fault diagnosis of gearbox based on EEMD and ICS-SVM［J］.Electric Power Science and Engineering，2020，36(10):56-62.(in Chinese)

[1]	段浩，陈晖，翟兆阳，韩雨，马凡华，崔亚辉. 基于支持向量机的氢混天然气发动机性能预测[J]. 兵工学报, 2022, 43(5): 1002-1011.
[2]	杨建新，兰小平，冯亚东，杨一铭，郭志明. 基于改进樽海鞘群和最小二乘支持向量机算法的新型弹药质量评估方法[J]. 兵工学报, 2022, 43(5): 1012-1022.
[3]	李志农, 胡志峰, 毛清华, 张旭辉, 陶俊勇. 非线性调频模态分解-同步提取变换方法及其在滚动轴承故障诊断中的应用[J]. 兵工学报, 2021, 42(6): 1324-1330.
[4]	朱敏，许爱强，许晴，李睿峰. 基于改进多层核超限学习机的模拟电路故障诊断[J]. 兵工学报, 2021, 42(2): 356-369.
[5]	马润波，董丽虹，王海斗. 基于支持向量机的热喷涂层接触疲劳寿命预测[J]. 兵工学报, 2021, 42(12): 2743-2752.
[6]	雷震烁，刘松涛，温镇铭，葛杨. 复杂电磁环境下末制导雷达工作状态实时判别方法[J]. 兵工学报, 2021, 42(1): 83-90.
[7]	王成，吴岩，杨廷飞. 利用改进单分类支持向量机提升舰船尾流目标的检测准确率[J]. 兵工学报, 2020, 41(9): 1887-1893.
[8]	吕国俊，曹建军，郑奇斌，常宸，翁年凤，彭琮. 基于多目标蚁群优化的单类支持向量机相似重复记录检测[J]. 兵工学报, 2020, 41(2): 324-331.
[9]	邓青，薛青，罗佳. 基于支持向量机的坦克驾驶模拟训练结果分析[J]. 兵工学报, 2019, 40(9): 1953-1960.
[10]	李佳蔚，崔涛，刘宇航，师帅楠，孙强. 基于排气温度动态模型的在线观测器研究[J]. 兵工学报, 2019, 40(8): 1562-1571.
[11]	代健，李泽，郝新红，栗苹. 基于目标联合特征提取的脉冲多普勒引信抗干扰方法[J]. 兵工学报, 2019, 40(2): 225-233.
[12]	郭伟超，赵怀山，李成，李言，汤奥斐. 基于小波包能量谱与主成分分析的轴承故障特征增强诊断方法[J]. 兵工学报, 2019, 40(11): 2370-2377.
[13]	翟禹尧，史贤俊，吕佳朋. 基于广义随机Petri网的导弹系统测试性建模与指标评估方法研究[J]. 兵工学报, 2019, 40(10): 2070-2079.
[14]	张伟，刘星，许爱强，平殿发. l_p范数约束的模拟电路3层多核故障诊断模型[J]. 兵工学报, 2018, 39(7): 1352-1363.
[15]	于震梁, 孙志礼, 曹汝男, 王鹏. 基于支持向量机和卡尔曼滤波的机械零件剩余寿命预测模型研究[J]. 兵工学报, 2018, 39(5): 991-997.