Development and validation of a machine learning-driven predictive combustion model for a hydrogen-enriched compressed natural gas engine

LI Hongzhi; HU Zhenyu; TAO Weijun; HUANG Yongcheng

doi:1

您当前的位置：

首页 >

文章列表页 >

Development and validation of a machine learning-driven predictive combustion model for a hydrogen-enriched compressed natural gas engine

更新时间：2025-11-18

- Development and validation of a machine learning-driven predictive combustion model for a hydrogen-enriched compressed natural gas engine
- Moren Journal (2025)
- 作者机构：
  
  1.西安交通大学能源与动力工程学院,陕西省西安市710049
  2.比亚迪汽车工业有限公司,广东省深圳市518118
- 作者简介：
- 基金信息：
- DOI：1
  CLC： TK432
- Received：22 April 2025，
  
  Revised：2025-07-03，
  
  Accepted：05 July 2025，
- 稿件说明：
移动端阅览
LI Hongzhi, HU Zhenyu, TAO Weijun, et al. Development and validation of a machine learning-driven predictive combustion model for a hydrogen-enriched compressed natural gas engine[J/OL]. Moren Journal, 2025.
DOI：

LI Hongzhi, HU Zhenyu, TAO Weijun, et al. Development and validation of a machine learning-driven predictive combustion model for a hydrogen-enriched compressed natural gas engine[J/OL]. Moren Journal, 2025. DOI： 1.

摘要

针对一维性能模拟软件中缺少天然气掺氢（HCNG）燃烧基础数据的问题，同时为准确模拟HCNG特殊物化特性对发动机性能的影响，本研究开发了一种新的机器学习驱动的可预测燃烧模型。首先基于人工神经网络（ANN）的机器学习方法开发了HCNG的层流燃烧速度和着火延迟时间计算模型，然后根据试验数据对新开发的可预测燃烧模型进行标定和验证，最后在外特性上，利用所开发的模型研究了掺氢比对发动机性能的影响。结果表明：经训练和验证后的层流燃烧速度和着火延迟时间计算模型相关系数超过0.99；利用标定后模型计算得到的发动机性能参数的误差小于3%，燃烧相位角和爆震限制点火提前角的误差小于2 °CA；受爆震和涡前排温的限制，最大允许掺氢比为20%，且随着掺氢比的增大，最佳点火提前角逐渐推迟，有效热效率随之下降；掺氢可加快燃烧但会增加爆震倾向，而废气再循环（EGR）虽然会减缓燃烧但可有效抑制爆震、提前燃烧相位，两者协同控制可提高发动机的有效热效率。

Abstract

To address the lack of fundamental combustion data for Hydrogen-enriched Compressed Natural Gas (HCNG) in one-dimensional performance simulation software

and to accurately simulate the effects of HCNG’s special physical and chemical properties on engine performance

a novel machine learning-driven predictive combustion model was developed in this paper. Firstly

a machine learning approach based on artificial neural networks (ANN) was employed to develop computational models for the laminar burning velocity and ignition delay time of HCNG. Then

the newly developed combustion model was calibrated and validated against experimental data. Finally

the validated model was applied to investigate the effects of hydrogen blending ratio on engine performance under full-load conditions. The results show that the developed laminar burning velocity and ignition delay time models achieved correlation coefficients exceeding 0.99 after training and validation. Errors in predicted engine performance parameters using the calibrated combustion model were less than 3%

while errors in predicted combustion phasing angle and knock-limited spark advance angle were within 2 °CA. The maximum allowable hydrogen blending ratio is limited to 20% due to knock and pre-turbine exhaust temperature constraints. As the hydrogen blending ratio increases

the optimal spark timing gradually retards

leading to a decrease in effective thermal efficiency. Additionally

hydrogen blending accelerates combustion but increases knock propensity

while exhaust gas recirculation (EGR) slows combustion but effectively suppresses knock and advances combustion phasing. Synergistic control of hydrogen blending and EGR can enhance the engine’s effective thermal efficiency.

关键词

Keywords

references

彭天铎，袁志逸，任磊，等 . 中国碳中和目标下交通部门低碳发展路径研究［J］. 汽车工程学报， 2022 , 12 ( 4 ): 351 - 359 .

PENG Tianduo ， YUAN Zhiyi ， REN Lei ， et al . Pathway for China’s Transport Sector Towards Carbon Neutrality Target ［J］. Chinese Journal of Automotive Engineering ， 2022 , 12 ( 4 ): 351 - 359 .

ZHANG W , WANG Y , LONG W , et al . Numerical investigation on knock intensity, combustion, and emissions of a heavy-duty natural gas engine with different hydrogen mixing strategies [J ] . International Journal of Hydrogen Energy , 2024 , 62 : 551 - 561 .

CHEN L , ZHANG X , ZHANG R , et al . Effect of hydrogen direct injection on natural gas/hydrogen engine performance under high compression-ratio conditions [J ] . International Journal of Hydrogen Energy , 2022 , 47 ( 77 ): 33082 - 33093 .

ALGAYYIM S J M , SALEH K , WANDEL A P , et al . Influence of natural gas and hydrogen properties on internal combustion engine performance, combustion, and emissions: A review [J ] . Fuel , 2024 , 362 : 130844 .

DE SIMIO L , IANNACCONE S , GUIDO C , et al . Natural Gas/Hydrogen blends for heavy-duty spark ignition engines: Performance and emissions analysis [J ] . International Journal of Hydrogen Energy , 2024 , 50 : 743 - 757 .

BARATTA M , CHIRICHES S , GOEL P , et al . CFD modelling of natural gas combustion in IC engines under different EGR dilution and H2-doping conditions [J ] . Transportation Engineering , 2020 , 2 : 100018 .

吴长水，李聪，陈凌珊．天然气掺氢发动机的性能研究［J］．小型内燃机与摩托车， 2011 ， 40 ( 3 )： 71 - 73 ．

WU Changshui , LI Chong , CHEN Lingshan . Research on the performance of hydrogen-doped natural gas engine [J ] . Small Internal Combustion Engine and Motorcycle , 2011 , 40 ( 3 ): 71 - 73 .

董永超，马志豪，王鑫，等．掺氢比和废气再循环率对天然气发动机性能的影响［J］．河南科技大学学报(自然科学版) ， 2021 ， 42 ( 2 ) : 30 - 37 ．

DONG Y C ， MA Z H ， WANG X ， et al ． Effects of hydrogen ratio and EGR rate on the performance of compressed natural gas engine ［J］． Journal of Henan university of science and technology( natural science) ， 2021 ， 42 ( 2 ) : 30 - 37 ．

RAO A , WU Z , KUMAR MEHRA R , et al . Effect of hydrogen addition on combustion, performance and emission of stoichiometric compressed natural gas fueled internal combustion engine along with exhaust gas recirculation at low, half and high load conditions [J ] . Fuel , 2021 , 304 : 121358 .

SHIVAPUJI A M , DASAPPA S . Quasi dimensional numerical investigation of syngas fuelled engine operation: MBT operation and parametric sensitivity analysis [J ] . Applied Thermal Engineering , 2017 , 124 : 911 - 928 .

SJERIĆ M , KOZARAC D , SCHUEMIE H , et al . A new quasi-dimensional flame tracking combustion model for spark ignition engines [J ] . Energy Conversion and Management , 2018 , 165 : 263 - 275 .

MA F , LI S , ZHAO J , et al . A fractal-based quasi-dimensional combustion model for SI engines fuelled by hydrogen enriched compressed natural gas [J ] . International Journal of Hydrogen Energy , 2012 , 37 ( 12 ): 9892 - 9901 .

PERINI F , PALTRINIERI F , MATTARELLI E . A quasi-dimensional combustion model for performance and emissions of SI engines running on hydrogen–methane blends [J ] . International Journal of Hydrogen Energy , 2010 , 35 ( 10 ): 4687 - 4701 .

KREBS S , BIET C . Predictive model of a premixed, lean hydrogen combustion for internal combustion engines [J ] . Transportation Engineering , 2021 , 5 : 100086 .

李昭．基于机器学习的C 1 -C 4 烷烃燃料层流燃烧速度预测［D］．西安：长安大学， 2021 ．

RAO A , CHEN T , LIU Y , et al . Computational analysis of performances for a hydrogen enriched compressed natural gas engine’ by advanced machine learning algorithms [J ] . Fuel , 2023 , 347 : 128244 .

FARHAN M , CHEN T , SHAHID MI , et al . Effect of exhaust gas recirculation on knock of HCNG fueled spark ignition engine and prediction of knock intensity by improved particle swarm optimization-back propagation neural network . Applied Thermal Engineering 2025 ; 264 : 125439 .

FARHAN M , CHEN T , RAO A , et al . Comparative knock analysis of HCNG fueled spark ignition engine using different heat transfer models and prediction of knock intensity by artificial neural network fitting tool . Energy 2024 ; 304 : 132135 .

AN J , HE G , LUO K , et al . Artificial neural network based chemical mechanisms for computationally efficient modeling of hydrogen/carbon monoxide/kerosene combustion [J ] . International Journal of Hydrogen Energy , 2020 , 45 ( 53 ): 29594 - 29605 .

SHIVAPUJI A M , DASAPPA S . Analysis of thermodynamic scope engine simulation model empirical coefficients: Suitability assessment and tuning of conventional hydrocarbon fuel coefficients for bio syngas [J ] . International Journal of Hydrogen Energy , 2017 , 42 ( 26 ): 16834 - 16854 .

张文嘉．基于K-k-ε流动模型与可预测燃烧模型的点燃式发动机性能模拟和开发［D］．西安：西安交通大学， 2018 .

FOGLA N , BYBEE M , MIRZAEIAN M , et al . Development of a K-k-ε Phenomenological Model to Predict In-Cylinder Turbulence [J ] . SAE International Journal of Engines , 2017 , 10 ( 2 ): 562 - 575 .

LIVENGOOD J. C. and WU P. C. , Correlation of autoignition phenomena in internal combustion engines and rapid compression machines , Symp . Int . Combust , 1955 , 5 : 347 - 356 .

SMITH G , Golden D , Frenklach M , et al . GRI-Mech 3.0 [EB/OL ] . 2006 . http://combustion.berkeley.edu/gri-mech/ http://combustion.berkeley.edu/gri-mech/ .

ZHOU C W , LI Y , O’CONNOR E , et al . A comprehensive experimental and modeling study of isobutene oxidation [J ] . Combustion and Flame , 2016 , 167 : 353 - 379 .

BERWAL P , SOLAGAR S , KUMAR S . Experimental investigations on laminar burning velocity variation of CH4+ H2+ air mixtures at elevated temperatures [J ] . International Journal of Hydrogen Energy , 2022 , 47 ( 37 ): 16686 - 16697 .

HUANG J , HILL P G , BUSHE W K , et al . Shock-tube study of methane ignition under engine-relevant conditions: experiments and modeling [J ] . Combustion and flame , 2004 , 136 ( 1-2 ): 25 - 42 .

HANN S . A quasi-dimensional SI burn rate model for carbon-neutral fuels [M ] . Springer Fachmedien Wiesbaden , 2021 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Predictive Combustion Model for Hydrogen-Enriched Compressed Natural Gas Engines Using Machine Learning

I/O Performance Modeling and Prediction Method for Large-Scale Parallel Applications Using Machine Learning

Automatic Classification of Pyrocarbon Texture under Polarized Light Microscope Based on Artificial Neural Network

Aerodynamic Optimization Design of Vaned Diffuser for Centrifugal Compressor under Stage Environment

Mechanism and Optimal Control Strategy for Hysteresis in Rapid Dehumidification Response of Indirect Thermal Systems in Electric Vehicles

Related Author

LI Hongzhi

HU Zhenyu

TAO Weijun

HUANG Yongcheng

DONG Xiaoshe

LONG Bingzhi

LIU Xindong

WANG Qiang

Related Institution

BYD Auto Industry Co., Ltd.

School of Energy and Power Engineering, Xi'an Jiaotong University

School of Software，Xi'an Jiaotong University

Network Information Center，Xi'an Jiaotong University

School of Computer Science and Technology，Xi'an Jiaotong University

⁰