基于CNN-Transformer融合框架的柴油车氨排放预测方法

Abstract
Figure/Table
References
Related Citation (15)

Download: PDF (1765 KB) HTML (1 KB)
Export: BibTeX | EndNote (RIS)

Abstract In this study, a diesel vehicle NH₃ emission prediction model based on the fusion framework of Convolutional Neural Network (CNN) and Transformer is proposed. The model was developed by integrating the local feature extraction capability of CNN with the global dependency modeling capability of Transformer, enabling the highly accurate prediction of NH₃ emissions from diesel vehicles under real road driving conditions. The study was conducted based on the actual on-road emissions test data of an N₃-class diesel vehicle. Feature screening was performed using the Pearson correlation coefficient method, and the key hyperparameters of the model were optimized through the application of the Bayesian algorithm, which enhanced its performance. Additionally, the SHapley Additive exPlanations (SHAP) algorithm was utilized to identify the pivotal factors influencing NH₃ emissions. The results indicated that the proposed model achieved highly accurate predictions of NH₃ emissions from diesel vehicles in real road driving conditions when tested on an independent dataset. The R², MAE, and MSE values of the predicted NH₃ concentration compared to the actual measured values were 0.986, 0.663, and 2.285, respectively, which were significantly superior to those obtained by the traditional Random Forest (RF) model, the Long Short-Term Memory (LSTM) neural network model, and the Transformer model. This study provided an efficient and reliable method for monitoring NH₃ emissions from in-use diesel vehicles and offered a novel perspective for elucidating the principal factors influencing NH₃ emissions from diesel vehicles on the road.

Key words： diesel vehicles emission NH₃ convolutional neural network transformer

Received: 20 August 2024

PACS:

X511

Corresponding Authors: 白晓鑫,工程师,baixiaoxin@catarc.ac.cn E-mail: baixiaoxin@catarc.ac.cn

	Service

	E-mail this article
	Add to my bookshelf
	Add to citation manager
	E-mail Alert
	RSS
	Articles by authors
	BAI Xiao-xin
	GUO Xiang-yang
	WU Chun-ling
	WANG Feng-bin
	LI Xu
	LIU Wei-lin

Cite this article:

BAI Xiao-xin,GUO Xiang-yang,WU Chun-ling等. Research on diesel vehicle NH₃ emission prediction method based on CNN-Transformer fusion framework[J]. CHINA ENVIRONMENTAL SCIENCECE, 2025, 45(3): 1231-1240.

URL:

http://www.zghjkx.com.cn/EN/ OR http://www.zghjkx.com.cn/EN/Y2025/V45/I3/1231

[1] 中国移动源环境管理年报(2023年)[J].环境保护, 2024,52(2):48-62. China mobile source environmental management annual report (2023)[J]. Environmental Protection, 2024,52(2):48-62.
[2] Anderson N, Strader R, Davidson C. Airborne reduced nitrogen:Ammonia emissions from agriculture and other sources[J]. Environment international, 2003,29(2/3):277-286.
[3] Zhang H, Ying Q. Source apportionment of airborne particulate matter in Southeast Texas using a source-oriented 3D air quality model[J]. Atmospheric Environment, 2010,44(29):3547-3557.
[4] Farren, N J., Davison J, Rose R A, et al. Underestimated ammonia emissions from road vehicles[J]. Environmental science& technology, 2020,54(24):15689-15697.
[5] 陈培林,肖欣欣,王勤耕.基于卫星观测的2010~2020年中国高分辨率NH₃排放特征[J].中国环境科学, 2023,43(6):2673-2682. Chen P L, Xiao X X, Wang Q G, et al. High-resolution characteristics of NH₃ emission from 2010 to 2020 in China based on satellite observation[J]. China Environmental Science, 2023,43(6):2673-2682.
[6] 白晓鑫,吴春玲,刘卫林,等.柴油车尿素溶液品质在线检测方法研究.汽车实用技术, 2024,49(12):89-94. Bai X X, Wu C L, Liu W L, et al. Research on online detection method for urea solution quality of diesel vehicle[J]. Automobile Applied Technology, 2024,49(12):89-94.
[7] GB 17691-2018重型柴油车污染物排放限值及测量方法(中国第六阶段)[S]. GB 17697-2018 Limits and measurement methods for emissions from diesel-fueled heavy-duty vehicles (China VI)[S].
[8] Suarez-Bertoa R, Mendoza-Villafuerte P, Riccobono F, et al. On-road measurement of NH₃ emissions from gasoline and diesel passenger cars during real world driving conditions[J]. Atmospheric Environment, 2017,166:488-497.
[9] Zhu H, Ma T, Toumasatos Z, et al. On-road NO_x and NH₃emissions measurements from in-use heavy-duty diesel and natural gas trucks in the South Coast air Basin of California[J]. Atmospheric Environment, 2024,316:120179.
[10] Mendoza-Villafuerte P, Suarez-Bertoa R, Giechaskiel B, et al. NO_x, NH₃, N₂O and PN real driving emissions from a Euro VI heavy-duty vehicle. Impact of regulatory on-road test conditions on emissions[J]. Science of the Total Environment, 2017,609:546-555.
[11] European Commission. Proposal for a regulation of the European parliament and of the council on type-approval of motor vehicles and of engines and of systems, components and separate technical units intended for such vehicles, with respect to their emissions and battery durability (Euro 7)[EB/OL].(2022-11-10)[2024-06-04] . https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX% 3A52022PC0586.
[12] Kousoulidou M, Georgios F, Leonidas N, et al. Use of portable emissions measurement system (PEMS) for the development and validation of passenger car emission factors[J]. Atmospheric Environment, 2013,64:329-338.
[13] Thiruvengadam A, Besch M C, Thiruvengadam P, et al. Emission rates of regulated pollutants from current technology heavy-duty diesel and natural gas goods movement vehicles[J]. Environmental Science& Technology, 2015,49(8):5236-5244.
[14] Huang C, Hu Q, Lou S, et al. Ammonia emission measurements for light-duty gasoline vehicles in China and implications for emission modeling[J]. Environmental Science& Technology, 2018,52(19):11223-11231.
[15] Pla B, Piqueras P, Bares P, et al. NO_x sensor cross sensitivity model and simultaneous prediction of NO_x and NH₃ slip from automotive catalytic converters under real driving conditions[J]. International Journal of Engine Research, 2021,22(10):3209-3218.
[16] Wen L, Li X, Gao L, et al. A new convolutional neural network-based data-driven fault diagnosis method[J]. IEEE Transactions on Industrial Electronics, 2017,65(7):5990-5998.
[17] Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need[J]. Advances in Neural Information Processing Systems, 2017:5998-6008.
[18] 黄茂庭,徐金明.使用CNN (卷积神经网络)-LSTM (长短期记忆)联合神经网络预测盾构隧道施工引起的地面沉降[J].城市轨道交通研究, 2024,27(6):166-171. Huang M T, Xu J M. Prediction of land subsidence caused by shield tunnel construction with joint CNN-LSTM neural network[J]. Urban Mass Transit, 2024,27(6):166-171.
[19] Thakkar V, Tewary S, Chakraborty C. Batch Normalization in convolutional neural networks-A comparative study with CIFAR-10 data[C]//2018fifth international conference on emerging applications of information technology (EAIT). IEEE, 2018:1-5.
[20] Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need[J]. Advances in Neural Information Processing Systems, 2017,30:5998-6008.
[21] Williamson D F, Parker R A, Kendrick J S. The box plot:a simple visual method to interpret data[J]. Annals of internal medicine, 1989, 110(11):916-921.
[22] 陈婉娇.缺失数据插补方法及其在医学领域的应用研究[D].广州:华南理工大学, 2020. Chen W J.Research on application of missing data lmputation in medical field[D].Guang Zhou:South China University of Technology, 2020.
[23] He L, Li G, Wu X, et al. Characteristics of NO_x and NH₃ emissions from in-use heavy-duty diesel vehicles with various aftertreatment technologies in China[J]. Journal of Hazardous Materials, 2024, 465:133073.
[24] Chen X, Liang C, Huang D, et al. Evolved optimizer for vision[C]//First Conference on Automated Machine Learning (Late-Breaking Workshop). 2022.
[25] Loshchilov I, Hutter F. SGDR:Stochastic gradient descent with warm restarts[J]. 2016.DOI:10.48550/arXiv.1608.03983.
[26] 白晓鑫,吴春玲,景晓军,等.基于RLS和BO算法的重型车载重估算研究[J].汽车实用技术, 2023,48(5):56-63. Bai X X, Wu C L, Jing X J, et al. Research on heavy-duty vehicle mass estimation based on recursive least square and bayesian optimization algorithm[J]. Automobile Applied Technology, 2023, 48(5):56-63.
[27] Jung Y. Multiple predicting K-fold cross-validation for model selection[J]. Journal of nonparametric statistics, 2018,30(1):197-215.
[28] Marcílio W E, Eler D M. From explanations to feature selection:assessing SHAP values as feature selection mechanism[C]//202033rd SIBGRAPI conference on Graphics, Patterns and Images (SIBGRAPI). IEEE, 2020:340-347.