基于Res⁃CNN和燃油压力波的柴油机喷油器故障诊断方法

doi:10.19562/j.chinasae.qcgc.2021.06.019

Abstract

Abstract:

The working quality of the fuel injection system influences the working process and performance of diesel engine directly. Due to the difficulty in recognizing the characteristic points in fuel pressure wave automatically by using fuel pressure wave for fault diagnosis，the on?line fault diagnosis is influenced. This paper proposes a fuel injector fault diagnosis method by deep learning image recognition theory. Experiments are conducted on a fuel injection pump test bench to simulate typical faults，and the fuel pressure wave of high?pressure fuel pipe is measured . The fuel pressure wave characteristics and laws under different fault conditions are analyzed. A deep residual CNN network （Res?CNN） model is built to detect and verify the faults，with the one?dimension fuel pressure wave signal as the input，and the learning process of fault characteristics is analyzed visually. The results show that the model has higher diagnostic accuracy than the traditional method， which verifies the feasibility of direct application of fuel pressure wave image recognition method for on?line real?time monitoring.

Key words: diesel engine, fuel injector, fault diagnosis, deep learning, convolutional neural network

Ying Jin,Xinyong Qiao,Cheng Gu,Hao Guo,Chuming Ning. A Method for Fault Diagnosis of Fuel Injector of Diesel Engine Based on Res⁃CNN and Fuel Pressure Wave[J].Automotive Engineering, 2021, 43(6): 943-951.

Figures/Tables 17

故障模式	$T a c'$	$T b d'$	$T a b'$	$p a b'$	$p a c'$	$p a d'$
M1	0.465	0.520	0.158	1.323	1.930	2.083
M2	0.496	0.581	0.332	1.200	1.289	1.785
M3	0.387	0.320	0.138	3.514	17.464	11.405
M4	0.498	0.592	0.185	1.037	5.887	0.242
M5	0.409	0.511	0.148	1.563	1.711	2.051
M6	0.481	0.526	0.156	1.549	3.704	1.586

模型结构名称	输出张量尺寸	网络层超参数（卷积核大小、个数、步长）
输入	1×2800	None
Conv1	1×700	1×7，64， stride 2
Conv1	1×700	1×3， max pooling，stride 2
Conv2	1×234	$1 × 1,64 1 × 5,64 1 × 1,256 × 1, s t r i d e ? 3$
Conv3	1×78	$1 × 1,128 1 × 5,128 1 × 1,512 × 1, s t r i d e ? 3$
Conv4	1×26	$1 × 1,256 1 × 5,256 1 × 1,1024 × 1, s t r i d e ? 3$
Conv5	1×7	$1 × 1,512 1 × 5,512 1 × 1,2048 × 1, s t r i d e ? 3$
全局平均池化	1×1×1000	None
输出层	6	损失函数：Softmax
模型参数量	18.02 MB

References 18

1	乔新勇，刘东利，康葳，等. 基于压力波的柴油机喷油器故障诊断研究［J］.无损检测，2004，26（4）：180-189.
	QIAO X Y， LIU D L， KANG W， et al. Research on fault diagnosis for diesel injector based on fuel pressure waves［J］. Nondestructive Testing， 2004，26（4）：180-189.
2	马超，王伏，龚笑舞，等.一种共轨喷油器故障诊断策略开发［J］.现代车用动力，2016，3：49-53.
	MA C， WANG F， GONG X W， et al. Development of a diagnostic strategy for common rail injector［J］. Modern Vehicle Power， 2016，3：49-53.
3	王金鑫，王忠巍，马修真，等. 柴油机燃油系统多故障的解耦与诊断技术［J］.控制与决策，2019， 34（10）：2249-2255.
	WANG J X， WANG Z W， MA X Z， et al. Decoupling and diagnosis of multi⁃fault of diesel engine fuel system［J］. Control and Decision， 2019， 34（10）：2249-2255.
4	孙宜权，王滨，张英堂，等. 基于自适应平行因子的柴油机喷油故障诊断研究［J］. 兵工学报，2013，34（5）：519-526.
	SUN Y Q， WANG B， ZHANG Y T， et al. Study of fault diagnosis of diesel engine fuel injection based on adaptive parallel factor［J］. Acta Armamentarii， 2013， 34（5）：519-526.
5	崔海英，石秀华，王文斌，等. 基于LMBP算法的柴油机喷油器故障诊断［J］.振动、测试与诊断，2009，29（4）：466-469，481.
	CUI H Y， SHI X H， WANG W B， et al. Fault diagnosis of diesel injector using levenberg⁃marquardt back propagation algorithm［J］. Journal of Vibration， Measurement and Diagnosis， 2009， 29（4）：466-469，481.
6	曹继平，王赛，岳小丹，等.基于自适应深度卷积神经网络的发射车滚动轴承故障诊断研究［J］.振动与冲击，2020，39（5）：97-104.
	CAO J P， WANG S， YUE X D， et al. Rolling bearing fault diagnosis of launch vehicle based on adaptive deep CNN［J］. Journal of Vibration and Shock，2020，39（5）：97-104.
7	朱丹宸，张永祥，潘洋洋，等.基于多传感器信号和卷积神经网络的滚动轴承故障诊断［J］.振动与冲击，2020，39（4）：172-178.
	ZHU D C， ZHANG Y X， PAN Y Y， et al. Fault diagnosis for rolling element bearings based on multi⁃sensor signals and CNN［J］.Journal of Vibration and Shock， 2020，39（4）：172-178.
8	宫文峰，陈辉，张美，等.基于深度学习的电机轴承微小故障智能诊断方法［J］.仪器仪表学报，2020，41（1）：195-205.
	GONG W F， CHEN H， ZHANG M， et al. Intelligent diagnosis method for incipient fault of motor bearing based on deep learning［J］. Chinese Journal of Scientific Instrument， 2020，41（1）：195-205.
9	黄鑫，陈仁祥，黄钰. 卷积神经网络在机械设备故障诊断领域应用与挑战［J］. 设计与研究， 2019（1）：96-100.
	HUANG X，CHEN R X，HUANG Y. Convolution neural networks for mechanical equipment fault diagnosis： the application and challenge［J］. Desing and Research， 2019（1）：96-100.
10	朱会杰，王新晴，芮挺，等. 基于平移不变CNN 的机械故障诊断研究［J］. 振动与冲击， 2019， 38（5）：45-52.
	ZHU H J，WANG X Q，RUI T，et al. Machinery fault diagnosis based on shift invariant CNN［J］. Journal of Vibration and Shock， 2019， 38（5）：45-52.
11	郁磊，史峰，王辉，等.MATLAB智能算法30个案例分析［M］.北京：北京航空航天大学出版社，2015：250-256.
	YU L， SHI F， WANG H， et al. Analysis of 30 cases of MATLAB intelligent algorithm［M］.Beijing： Beijing University of Aeronautics and Astronautics Press，2015： 250-256.
12	周兴康，余剑波. 基于深度一维残差卷积自编码网络的齿轮箱故障诊断［J］. 机械工程学报，2020，56（7）： 96-108.
	ZHOU X K， YU J B. Gearbox fault diagnosis based on one⁃dimension residual convolutional auto⁃encoder［J］. Journal of Mechanical Engineering， 2020，56（7）： 96-108.
13	魏秀参.解析深度学习：卷积神经网络原理与视觉实践［M］.北京：电子工业出版社，2018：32-33.
	WEI X C. Analytical deep learning： Principles of convolutional neural networks and visual practice［M］.Beijing：Electronic Industry Press，2018：32-33.
14	韩涛，袁建虎，唐建，等. 基于MWT 和CNN 的滚动轴承智能复合故障诊断方法［J］.机械传动， 2016，40（12）：139-143.
	HAN T， YUAN J H， TANG J， et al. An approach of intelligent compound fault diagnosis of rolling bearing based on MWT and CNN［J］. Journal Mechanical Transmission， 2016，40（12）：139-143.
15	吴魁，王仙勇，孙洁，等. 基于深度卷积网络的多传感器信号故障诊断方法研究［J］.计算机测量与控制，2018，26（1）：18-21.
	WU K， WANG X Y， SUN J， et al. Study of multi⁃sensor fault diagnose method based on convolutional neural networks［J］. Computer Measurement & Control， 2018，26（1）：18-21.
16	高清春，胡甫才.基于变分模态和奇异值分解的柴油机气门间隙故障诊断方法研究［J］.武汉理工大学学报，2019，43（4）：746-751.
	GAO Q C， HU P C. Research on the fault diagnosis method of diesel engine valve clearance based on vibrational mode and singular value decomposition［J］. Journal of Wuhan University of Technology， 2019，43（4）：746-751.
17	吴建波，王春艳，洪华军，等. 基于极限学习机的船舶柴油机故障诊断［J］.计算机工程与应用，2019， 55（15）：147-152.
	WU J B，WANG C Y，HONG H J， et al. Fault diagnosis of marine diesel engine based on extreme learning machine［J］. Computer Engineering and Applications， 2019， 55（15）：147-152.
18	YOO H J. Deep convolution neural networks in computer vision： A review ［J］. IEIE Transactions on Smart Processing and Computing，2015，4（1）：35-43.

[1]	SHEN Zhe, WANG Yi-Gang, YANG Zhi-Gang, LI Fang-Xu. [J]. , 2016, 38(12): 1440 -1445 .
[2]	ZENG Bi-Qiang, GAO Ji-Dong, PENG Wei, SUN Zhen-Dong. [J]. , 2016, 38(12): 1446 -1451 .
[3]	LIU Hui, WANG Xiao-Jie, XIANG Chang-Le. [J]. , 2016, 38(12): 1483 -1487 .
[4]	CHEN Wu-Wei, DENG Shu-Chao, HUANG He, XIE You-Hao. [J]. , 2016, 38(12): 1488 -1493 .
[5]	PENG Qian-Lei, GUI Liang-Jin, FAN Zi-Jie. [J]. , 2016, 38(12): 1500 -1507 .
[6]	DONG Zhu-Rong, ZHANG Xin, HU Song-Hua, QIU Hao. [J]. , 2017, 39(1): 79 -85 .
[7]	SHI Hai-Min, YU Xiao-Li, LU Guo-Dong, HUANG Yu-Qi, LIU Zhen-Tao, HUANG Rui. [J]. , 2017, 39(1): 102 -106 .
[8]	QIN Chao-Ju, YANG Zhen-Zhong, ZHANG Wei-Zheng, SONG Li-Ye, YUAN Yan-Peng. [J]. , 2017, 39(2): 133 -137 .
[9]	ZHANG Guan-Jun, ZHAO Xin-Feng, CAO Li-Bo. [J]. , 2017, 39(2): 150 -158 .
[10]	CAO Li-Bo, HU Yuan, YAN Ling-Bo, PENG Yu, SHI Xiang-南. [J]. , 2017, 39(2): 174 -180 .

网络模型	PNN	CNN	Res?CNN
识别准确率/%	77.5	95.24	100
训练时间/s	6.52	67.27	70.22

[1]	Ze Gao, Zunkang Chu, Jiasheng Shi, Fu Lin, Weixiong Rao, Haiyan Yu. Research on Fast Prediction Method of Stress Field of Automotive Parts Based on Graph Network [J]. Automotive Engineering, 2024, 46(1): 170-178.
[2]	Pengbo Zhang, Renxiang Chen, Yiming Shao, Shizheng Sun, Kaibo Yan. Research Review of Fault Diagnosis for Electric Drive Powertrain System of Pure Electric Vehicles [J]. Automotive Engineering, 2024, 46(1): 61-74.
[3]	Weiguo Liu,Zhiyu Xiang,Rui Liu,Guodong Li,Zixu Wang. Research on End-to-End Vehicle Motion Planning Method Based on Deep Learning [J]. Automotive Engineering, 2023, 45(8): 1343-1352.
[4]	Yixiao Zhang,Xiao Ma,Xinhui Lu,Zhi Wang,Weilin Zhuge,Shijin Shuai. An Investigation on Intake Preheating Characteristics Based on Spray Wall Impinging Ignition for Diesel Engine [J]. Automotive Engineering, 2023, 45(8): 1489-1498.
[5]	Lei Zhang, Keren Guan, Xiaolin Ding, Pengyu Guo, Zhenpo Wang, Fengchun Sun. Tire-Road Friction Estimation Method Based on Image Recognition and Dynamics Fusion [J]. Automotive Engineering, 2023, 45(7): 1222-1234.
[6]	Jiahao Zhao,Zhiquan Qi,Zhifeng Qi,Hao Wang,Lei He. Calculation of Heading Angle of Parallel Large Vehicle Based on Tire Feature Points [J]. Automotive Engineering, 2023, 45(6): 1031-1039.
[7]	Xia Zhao,Zhao Li,Rui Fu,Zhenzhen Ge,Chang Wang. Real-Time Detection of Distracted Driving Behavior Based on Deep Convolution-Tokens Dimensionality Reduction Optimized Visual Transformer Model [J]. Automotive Engineering, 2023, 45(6): 974-988.
[8]	Yanyan Chen,Hai Wang,Yingfeng Cai,Long Chen,Yicheng Li. Efficient Automatic Driving Instance Segmentation Method Based on Detection [J]. Automotive Engineering, 2023, 45(4): 541-550.
[9]	Fengchong Lan,Jikai Chen,Jiqing Chen,Xinping Jiang,Zihan Li,Wei Pan. Research on Lithium Battery Remaining Useful Life Prediction Method Driven by Real Vehicle Data [J]. Automotive Engineering, 2023, 45(2): 175-182.
[10]	Linhui Li,Xinliang Zhang,Yifan Fu,Jing Lian,Jiaxu Ma. Research on Visible Light and Infrared Post-Fusion Detection Based on TC-YOLOv7 Algorithm [J]. Automotive Engineering, 2023, 45(12): 2280-2290.
[11]	Xiaojun Zhang,Jingzhe Xi,Yanlei Shi,Anlu Yuan. Lightweight YOLOv7-R Algorithm for Road-Side View Target Detection [J]. Automotive Engineering, 2023, 45(10): 1833-1844.
[12]	Jichao Hong,Fengwei Liang,Haixu Yang,Kerui Li. Research on Intelligent Safety Management and Control Methods for Big-data-driven Battery Systems [J]. Automotive Engineering, 2023, 45(10): 1845-1861.
[13]	Jie Hu,Yuanjie Li,Hao Geng,Huangzheng Geng,Xiong Guo,Hongwei Yi. Construction of Vehicle Fault Knowledge Graph Based on Deep Learning [J]. Automotive Engineering, 2023, 45(1): 52-60.
[14]	Shaohua Liu,Haochuan Dong,Lizhong Shen. Study on Construction of Diesel Combustion Mechanism Based on Soot Particle Size Distribution Prediction and the Impact of Oxygen Concentration [J]. Automotive Engineering, 2023, 45(1): 93-103.
[15]	Jie Hu,Boyuan Xu,Zongquan Xiong,Minjie Chang,Di Guo,Lihao Xie. Cross-Domain Object Detection Algorithm Based on Multi-scale Mask Classification Domain Adaptive Network [J]. Automotive Engineering, 2022, 44(9): 1327-1338.

A Method for Fault Diagnosis of Fuel Injector of Diesel Engine Based on Res⁃CNN and Fuel Pressure Wave

RichHTML

PDF (PC)

Like

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 17

References 18

Related Articles 15

Metrics

Comments

Recommended 10

模型结构名称	输出张量尺寸	网络层超参数（卷积核大小、个数、步长）
输入	1×2 800	None
Conv1	1×560	1×5，32，stride 3
Conv1	1×560	1×3，最大池化，stride 2
Conv2	1×112	1×3，64，stride 3
Conv2	1×112	1×3，最大池化，stride 2
Conv3	1×28	1×2，64，stride 2
Conv3	1×28	1×2，最大池化，stride 2
全连接层	1×64	激活函数：ReLU
输出层	6	损失函数：Softmax
模型参数量	74.06 MB