A comparative investigation of advanced machine learning methods for predicting transient emission characteristic of diesel engine

Aug 3, 2026

Upload by Patrick

Machine learning offers an efficient and robust approach for predicting the transient emission characteristics of diesel engines. The study compares seven methods—ANN, SVM, NARX, LSTM, GRU, Transformer, and TCN—trained, validated, and tested using WHTC and WHSC cycle data, with R2, MAE, and RMSE as evaluation metrics. Input–output relationships are examined via Pearson and Spearman correlation, selecting the top six parameters as model inputs. Hyperparameters are optimized with GA and PSO to determine the best structures.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages67

Reading Time2 h 49 min

AccessFree to download

Highlights

Model comparison across emission components

Correlation-based feature selection

Hyperparameter optimization and recommendations

University of Birmingham A comparative investigation of advanced machine learning methods for predicting transient emission characteristic of diesel engine Liao, Jianxiong; Hu, Jie; Yan, Fuwu; Chen, Peng; Zhu, Lei; Zhou, Quan; Xu, Hongming; Li, Ji DOI: 10.1016/j.fuel.2023.128767 License: Creative Commons: Attribution-NonCommercial-NoDerivs (CC BY-NC-ND) Document Version Peer reviewed version Citation for published version (Harvard): Liao, J, Hu, J, Yan, F, Chen, P, Zhu, L, Zhou, Q, Xu, H & Li, J 2023, 'A comparative investigation of advanced machine learning methods for predicting transient emission characteristic of diesel engine', Fuel, vol. 350, 128767. [https://doi.org/10.1016/j.fuel.2023.128767](https://doi.org/10.1016/j.fuel.2023.128767) Link to publication on Research at Birmingham portal General rights Unless a licence is specified above, all rights (including copyright and moral rights) in this document are retained by the authors and/or the copyright holders. The express permission of the copyright holder must be obtained for any use of this material other than for purposes permitted by law. •Users may freely distribute the URL that is used to identify this publication. •Users may download and/or print one copy of the publication from the University of Birmingham research portal for the purpose of private study or non-commercial research. •User may use extracts from the document in line with the concept of ‘fair dealing’ under the Copyright, Designs and Patents Act 1988 (?) •Users may not further distribute the material nor use it for the purposes of commercial gain. Where a licence is displayed above, please note the terms and conditions of the licence govern your use of this document. When citing, please reference the published version. Take down policy While the University of Birmingham exercises care and attention in making items available there are rare occasions when an item has been uploaded in error or has been deemed to be commercially or otherwise sensitive. If you believe that this is the case for this document, [please contact UBIRA@lists.bham.ac.uk](please contact UBIRA@lists.bham.ac.uk) providing details and we will remove access to the work immediately and investigate. Download date: 02. Aug. 2026 Revised Manuscript clean and final version Click here to view linked References 1 A comparative investigation of advanced machine learning methods 2 for predicting transient emission characteristic of diesel engine 3 Jianxiong Liaoa,b,c, Jie Hua,b,c,*, Fuwu Yana,b,c, Peng Chene, Lei Zhue, Quan Zhoud, 4 Hongming Xud, Ji Lid 5 a Hubei Key Laboratory of Advanced Technology for Automotive Components, 6 Wuhan University of Technology, Wuhan 430070, PR China 7 b Hubei Collaborative Innovation Center for Automotive Components Technology, 8 Wuhan University of Technology, Wuhan 430070, PR China 9 cHubei Research Center for New Energy & Intelligent Connected Vehicle, Wuhan 10 University of Technology, Wuhan 430070, China 11 d School of Engineering, University of Birmingham, Birmingham, B15 2TT, UK 12 e Center of Research and Department, Kailong High Technology Company Limited, 13 Wuxi 214153, China 14 * [Corresponding author: auto_hj@163.com](Corresponding author: auto_hj@163.com) (Jie Hu) 15 Highlights 16 􀁺 Seven advanced machine learning methods applied to the transient emission 17 characteristic prediction of diesel engine were introduced and compared. 18 􀁺 The correlation between input parameters and emission characteristic parameters 19 was analyzed based on Pearson and Spearman correlation analysis methods. 20 􀁺 The optimal hyperparameters for each machine learning method were obtained 21 using GA and PSO algorithms. 22 􀁺 A hybrid prediction model that combines multiple suitable algorithms was 23 proposed to have excellent performance in all emission characteristic prediction. 24 Graphic abstract 25 26 Abstract

FAQs

Which machine learning methods are compared for transient diesel engine emission prediction?

The document compares ANN, SVM, NARX, LSTM, GRU, Transformer, and TCN, assessing their performance across emission characteristic predictions.

How are models trained and evaluated in the study?

Models are trained, validated, and tested using WHTC and WHSC cycles, and performance is measured using R2, MAE, and RMSE.

What approach is used to select input parameters for the models?

The study analyzes correlations between inputs and emission outputs using Pearson and Spearman correlation, then selects the top six important parameters as inputs.

Aug 3, 2026

Upload by Patrick