Journal of Jilin University(Engineering and Technology Edition) ›› 2021, Vol. 51 ›› Issue (5): 1565-1574.doi: 10.13229/j.cnki.jdxbgxb20200603

Previous Articles    

Prediction of gasoline engine steady state exhaust based on model group prediction method

Tao CHEN1(),Jing QIN1,2(),Hua ZHAO1,Qing-peng SU1,3,Yong LYU3,Kai ZHONG1,Ying-bo WANG1,Yi-qiang PEI1   

  1. 1.State Key Laboratory of Engines，Tianjin University，Tianjin 300072，China
    2.Internal Combustion Engine Research Institute，Tianjin University，Tianjin 300072，China
    3.GAC Automotive Research & Development Center，Guangzhou 511434，China
  • Received:2020-08-10 Online:2021-09-01 Published:2021-09-16
  • Contact: Jing QIN E-mail:chent_666@tju.edu.cn;qinjing@tju.edu.cn

RICH HTML

 9   

PDF (PC)

146

Abstract:

This paper aims to reduce the workload of gasoline engine bench test and improve the experimental efficiency by numerical simulation technology. The model group prediction method （optimized artificial neural network method） was used to model and predict the steady-state emissions of NOx， CO and HC in the process of gasoline engine bench test. The results showed that the model group prediction method has high reliability and may improve the accuracy of prediction results compared with the traditional single model prediction method. The training data set of neural network modeling may be appropriately reduced by using the method of separating points， and the better prediction ability may still be maintained. In the process of project development， only 30% of the test quantities were needed， and the test results may be used for the training of neural network model， which may better predict the emission of the remaining working conditions. Through the validation analysis of other models， the model group prediction method had a good universality in the prediction process of engine steady-state exhaust.

Key words: gasoline, neural networks, prediction, engine exhausts, bench test

CLC Number: 

  • TK411

Table 1

Structure parameters of verify model"

类型排量/L压缩比缸数缸径×行程/mm
汽油机A1.609.50479.7×81.1
汽油机B1.809.60482.5×84.1
汽油机C2.0011.65482.5×92.8
柴油机A7.7917.506115×125
柴油机B9.0017.006123×132

Table 2

Performance parameters of verify model"

类型最大功率/kW

最大功率转

速/（r·min-1）

最大转矩/（N·m）最大扭矩转速/（r·min-1）
汽油机A14056002402400~5200
汽油机B1185000~62002501500~4500
汽油机C1475100~60002801700~5000
柴油机A21924009981400
柴油机B294190019001100~1450

Fig.1

Neural network structure diagram"

Fig.2

Predict flow chart of single training optimal model"

Fig.3

Flow chart of model group predicting method"

Table 3

Selection of model training data by interval point method"

隔点取点法数据分类转速/（r·min-1）
间隔一组转速选取训练数据

训练组

转速选取

1000、1500、2000、2800、 3600、4500、5500

预测组

转速选取

1250、1750、2400、3200、4000、4500、5000
间隔两组转速选取训练数据

训练组

转速选取

1000、1750、2800、4000、5500

预测组

转速选取

1250、1500、2000、2400、3200、3600、4500、5000
间隔三组转速选取训练数据

训练组

转速选取

1000、2000、3600、5500

预测组

转速选取

1250、1500、1750、2400、2800、3200、4000、4500、5000

Table 4

Comparison of judgement coefficients of NOx emission modeling and prediction by different neural network methods"

项目传统BP神经网络方法训练最优 模型预测模型群预测法
建模0.988 230.989 790.95~0.983 7
预测0.953 750.968 530.983 59

Table 5

Comparison of judgement coefficients of CO emission modeling and prediction by different neural network methods"

项目传统BP神经网络方法

训练最优

模型预测

模型群预测法
建模0.963 890.964 860.95~0.952 6
预测0.822 910.791 230.959 45

Table 6

Comparison of judgement coefficients of HC emission modeling and prediction by different neural network methods"

项目传统BP神经网络方法

训练最优

模型预测

模型群预测法
建模0.952 140.972 030.95~0.957 0
预测0.934 290.793 060.960 72

Fig.4

Prediction results of emissions by different neural network methods"

Table 7

Prediction of individual models in model group on emission under operating conditions"

项目NOxCOHC
模型群中模型个数2000200309
优于模型群预测法预测结果的模型个数200
优于模型群预测法预测结果的模型占比/%0.1000

Fig.5

Comparison of prediction effect between single model prediction method and model group prediction method"

Table 8

Judgement coefficient of single model in model group for emission prediction under operating conditions"

项目NOxCOHC
A组数据建模预测0.976 520.936 090.954 50
B组数据建模预测0.972 700.931 230.955 89

Table 9

Universality analysis of model group prediction method in emission prediction of various models types"

项目NOxCOHC
汽油机A建模Rtrain20.9~0.927 40.9~0.995 50.9~0.967 5
汽油机A预测Rpredict20.827 890.989 340.868 72
该预测结果优于单个模型预测结果的概率（汽油机A）/%95.7095.7095.20
汽油机B建模Rtrain20.8~0.851 40.8~0.898 60.75~0.792 1
汽油机B预测Rpredict20.806 620.861 880.534 50
该预测结果优于单个模型预测结果的概率（汽油机B）/%99.6597.1075.45
汽油机C建模Rtrain20.9~0.993 70.9~0.990 10.9~0.984 5
汽油机C预测Rpredict20.842 110.635 770.893 86
该预测结果优于单个模型预测结果的概率（汽油机C）/%79.10100.0098.95
柴油机A建模Rtrain20.9~0.999 90.9~1.00.9~1.0
柴油机A预测Rpredict20.972 610.955 170.992 18
该预测结果优于单个模型预测结果的概率（柴油机A）/%98.6590.7099.68
柴油机B建模Rtrain20.9~0.991 3—0.9~0.996 9
柴油机B预测Rpredict20.969 19—0.945 68
该预测结果优于单个模型预测结果的概率（柴油机B）/%83.15—86.75
1 裴毅强, 张建业, 秦静, 等. 增压直喷汽油机起动怠速及混合气浓度对微粒排放的影响[J]. 天津大学学报: 自然科学与工程技术版, 2014, 47(10): 892-897.
Pei Yi-qiang, Zhang Jian-ye, Qin Jing, et al. Effect of starting idling condition and mixture concentration of a turbocharged GDI engine on particle emission [J]. Journal of Tianjin University(Science and Technology), 2014, 47(10): 892-897.
2 谢宗法, 付文超, 魏枫展, 等. 进气门早关对汽油机进气过程及燃烧性能的影响[J]. 吉林大学学报: 工学版, 2020, 50(3): 850-858.
Xie Zong-fa, Fu Wen-chao, Wei Feng-zhan, et al. Effect of early intake valve closing on intake process and combustion performance of SI engine [J]. Journal of Jilin University(Engineering and Technology Edition), 2020, 50(3): 850-858.
3 于秀敏, 商震, 张岳韬, 等. 废气再循环对直喷汽油机燃烧及排放影响的仿真[J]. 吉林大学学报: 工学版, 2016, 46(4): 1109-1117.
Yu Xiu-min, Shang Zhen, Zhang Yue-tao, et al. Simulation of EGR on combustion and emission of gasline direct-injection engine[J]. Journal of Jilin University(Engineering and Technology Edition), 2016, 46(4): 1109-1117.
4 孙平, 曹智, 于秀敏, 等. 直喷时刻对复合喷射汽油机暖机过程燃烧和排放的影响[J]. 吉林大学学报: 工学版, 2020, 50(6): 1941-1949.
Sun Ping, Cao Zhi, Yu Xiu-min, et al. Experimental study on effect of compound injection on cold start combustion and emission of gasoline engine[J]. Journal of Jilin University(Engineering and Technology Edition), 2020, 50(6): 1941-1949.
5 卫海桥, 裴自刚, 冯登全, 等. 压电喷油器多次喷射对GDI汽油机颗粒物排放的影响[J]. 吉林大学学报: 工学版, 2018, 48(1): 166-173.
Wei Hai-qiao, Pei Zi-gang, Feng Deng-quan,et al. Effect of multi-injection piezo injector on particulate emission in gasoline direct injection engine[J]. Journal of Jilin University(Engineering and Technology Edition), 2018, 48(1): 166-173.
6 Heinz-Jakob N, Jörn K, Richard D, et al. Die euro-6-motoren des modularen dieselbaukastens von volkswagen[J]. MTZ-Motortechnische Zeitschrift, 2013, 74(6): 440-447.
7 Raidt B. Local gaussian process regression in order to model air charge of turbocharged gasoline SI engines[C]∥SAE Technical Paper, 2016-01-0624.
8 Gutjahr T, Kleinegraeber H, Huber T, et al. Commercial vehicle engineering congress[J]. SAE Technical Paper, 2015-01-2796.
9 Lasko T A. Efficient inference of gaussian process modulated renewal processes with application to medical event data[J]. Uncertain Artif Intell, 2014(2014): 469-476.
10 Bernhard S. Introduction to Gaussian Processes[M]. Massachusetts: MIT Press, 2008.
11 Mackay D J C. Bayesian interpolation[J]. Neural Computation, 1992, 4(3): 415-447.
12 Kuzin D, Yang L, Isupova O, et al. Ensemble kalman filtering for online gaussian process regression and learning[C]∥21st International Conference on Information Fusion, Cambridge, UK, 2018: 39-46.
13 Kauermann G, Bove D, Sabanes H L. Objective bayesian model selection in generalized additive models with penalized splines[J]. Journal of Computational and Graphical Statistics, 2015, 24(2): 394-415.
14 Baek J, Adams A, Dolson J. Lattice-based high-dimensional gaussian filtering and the permutohedral lattice[J]. Journal of Mathematical Imaging & Vision, 2013, 46(2): 211-237.
15 Rot I, Daniel F P, Rinderknecht S. Investigation of black box modeling approaches for representation of transient gearshift processes in automotive powertrains with automatic transmission[C]∥SAE Technical Papers, 2015-01-1143.
16 Mao G, Zhang C, Shi K, et al. Prediction of the performance and exhaust emissions of ethanol-diesel engine using different neural network[J]. Energy Sources Part A Recovery Utilization and Environmental Effects, 2019, 104: 1656307.
17 Nestor L S Y. Modelling the infiltration process with a multi-layer perceptron artificial neural network[J]. Hydrological Sciences Journal, 2006, 51(1): 3-20.
18 李健. 基于深度学习的变循环发动机气路故障诊断[D]. 上海: 上海交通大学电子信息与电气工程学院, 2019.
Li Jian. Fault diagnosis for variable cycle engine gas path: a deep learning approach[D]. Shanghai: School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University, 2019.
19 肖勇, 赵云, 涂治东, 等. 基于改进的皮尔逊相关系数的低压配电网拓扑结构校验方法[J]. 电力系统保护与控制, 2019, 47(11): 37-43.
Xiao Yong, Zhao Yun, Tu Zhi-dong, et al. Topology checking method for low voltage distribution network based on improved pearson correlation coefficient[J]. Power System Protection and Control, 2019, 47(11): 37-43.
20 龚麒鉴, 郭亚宾, 陈焕新, 等. 基于粒子群优化算法和BP神经网络的变频压缩机功率预测[J]. 制冷学报, 2020, 41(1): 89-95.
Gong Qi-jian, Guo Ya-bin, Chen Huan-xin, et al. Prediction of variable-speed compressor power based on particle swarm optimization and back propagation neural network[J]. Journal of Refrigeration, 2020, 41(1): 89-95.
21 谢岩, 廖松地, 朱曼妮, 等. 轻型汽油车稳态工况下的尾气排放特征[J]. 环境科学, 2020, 41(7): 3112-3120.
Xie Yan, Liao Song-di, Zhu Man-ni, et al. Emission characteristics of light-duty gasoline vehicle exhaust based on acceleration simulation mode[J]. Environmental Science, 2020, 41(7): 3112-3120.
22 Jeeragal R, Subramanian K A. Experimental investigation for NOx emission reduction in hydrogen fueled spark ignition engine using spark timing retardation[J]. Journal of Thermal Science, 2019, 28(4): 789-800.
23 Kosmadakis G M, Rakopoulos D C, Rakopoulos C D. Methane/hydrogen fueling a spark-ignition engine for studying NO, CO and HC emissions with a research CFD code[J]. Fuel, 2016, 185(1): 903-915.
[1] Hou⁃jie LI,Fa⁃sheng WANG,Jian⁃jun HE,Yu ZHOU,Wei LI,Yu⁃xuan DOU. Pseudo sample regularization Faster R⁃CNN for traffic sign detection [J]. Journal of Jilin University(Engineering and Technology Edition), 2021, 51(4): 1251-1260.
[2] Zhi-jun LI,Hao LIU,Li-peng ZHANG,Zhen-guo LI,Yuan-kai SHAO,Zhi-yang LI. Simulation on influence of microstructure of the wall on deep bed filtration of particulate filter [J]. Journal of Jilin University(Engineering and Technology Edition), 2021, 51(2): 422-434.
[3] Xiao-hui WEI,Chang-bao ZHOU,Xiao-xian SHEN,Yuan-yuan LIU,Qun-chao TONG. Accelerating CALYPSO structure prediction with machine learning [J]. Journal of Jilin University(Engineering and Technology Edition), 2021, 51(2): 667-676.
[4] Chun-guo LIU,Xiao-tong YU,Tao YUE,Dong-lai LI,Ming-zhe ZHANG. Springback prediction for double-curvature stiffened panel during milling [J]. Journal of Jilin University(Engineering and Technology Edition), 2021, 51(1): 188-199.
[5] Wen-long MU,Jing-xin NA,Wei TAN,Guang-bin WANG,Hao SHEN,Jian-ze LUAN. Residual strength prediction of adhesive CFRP-aluminum alloy adhesively bonded joint based on FTIR analysis [J]. Journal of Jilin University(Engineering and Technology Edition), 2021, 51(1): 139-146.
[6] Hong-wei ZHAO,Xiao-han LIU,Yuan ZHANG,Li-li FAN,Man-li LONG,Xue-bai ZANG. Clothing classification algorithm based on landmark attention and channel attention [J]. Journal of Jilin University(Engineering and Technology Edition), 2020, 50(5): 1765-1770.
[7] Ji-qing CHEN,Qing-sheng LAN,Feng-chong LAN,Zhao-lin LIU. Trajectory tracking control based on tire force prediction and fitting [J]. Journal of Jilin University(Engineering and Technology Edition), 2020, 50(5): 1565-1573.
[8] Xue-ping FAN,Guang QU,Yue-fei LIU. Bridge extreme stress prediction based on new data assimilation algorithm [J]. Journal of Jilin University(Engineering and Technology Edition), 2020, 50(2): 572-580.
[9] Yu FANG,Li-jun SUN. Urban bridge performance decay model based on survival analysis [J]. Journal of Jilin University(Engineering and Technology Edition), 2020, 50(2): 557-564.
[10] Dan-tong OUYANG,Cong MA,Jing-pei LEI,Sha-sha FENG. Knowledge graph embedding with adaptive sampling [J]. Journal of Jilin University(Engineering and Technology Edition), 2020, 50(2): 685-691.
[11] Peng-hui WANG,Hong-xia QIAO,Qiong FENG,Hui CAO,Shao-yong WEN. Durability model of magnesium oxychloride-coated reinforced concrete under the two coupling factors [J]. Journal of Jilin University(Engineering and Technology Edition), 2020, 50(1): 191-201.
[12] Wen-ku SHI,Long CHEN,Gui-hui ZHANG,Zhi-yong CHEN. Modeling and tests for torsional characteristics of multi-stage stiffness dual mass flywheel torsional dampers [J]. Journal of Jilin University(Engineering and Technology Edition), 2020, 50(1): 44-52.
[13] Yuan-li GU, Yuan ZHANG, Xiao-ping RUI, Wen-qi LU, Meng LI, Shuo WANG. Short⁃term traffic flow prediction based on LSSVMoptimized by immune algorithm [J]. Journal of Jilin University(Engineering and Technology Edition), 2019, 49(6): 1852-1857.
[14] Wu⁃jiao XU,Cheng⁃shang LIU,Xin⁃yao LU. Simulation and prediction of surface roughness of 6061 aluminum alloy workpiece after shot peening [J]. Journal of Jilin University(Engineering and Technology Edition), 2019, 49(4): 1280-1287.
[15] Xing⁃ye WANG,Jin⁃qiu ZHANG,Guo⁃qiang LI,Zhi⁃zhao PENG. Influence of inertial mass on rack and pinion actuator′s damping characteristic [J]. Journal of Jilin University(Engineering and Technology Edition), 2019, 49(3): 881-887.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
Copyright © Journal of Jilin University(Engineering and Technology Edition), All Rights Reserved.
Tel: +86-431-85095297　Fax: +86-431-85094074　E-mail: xbgxb@jlu.edu.cn
Powered by Beijing Magtech Co. Ltd