上海交通大学学报

上海交通大学学报 >

2022 , Vol. 56 >Issue 5: 604 - 610

DOI: https://doi.org/10.16183/j.cnki.jsjtu.2021.231

机械与动力工程

基于外推高斯过程回归方法的发动机排放预测

展开
  • 上海交通大学 动力装置与自动化研究所; 海洋工程国家重点实验室,上海 200240
王子垚(1996-),男,河南省郑州市人,硕士生,主要从事发动机排放预测、高斯过程回归等研究.

收稿日期: 2021-06-29

  网络出版日期: 2022-06-07

收起

Engine Emission Prediction Based on Extrapolated Gaussian Process Regression Method

Expand
  • Institute of Marine Power Plant and Automation; State Key Laboratory of Ocean Engineering, Shanghai Jiao Tong University, Shanghai 200240, China

Received date: 2021-06-29

  Online published: 2022-06-07

Fold

摘要

为了提高训练集范围之外行驶工况的预测精度,提出外推高斯过程回归(GPR)方法.首先,采用训练集数据对GPR模型进行预训练,然后在正负3个标准差之间均匀采样构建宽域输入集,以该输入集的预测方差均值最小为目标优化GPR模型超参数.某直喷汽油机转毂试验的结果表明,外推GPR的平均绝对误差为0.53411,比传统GPR降低24.27%,比反向传播神经网络降低36.32%.所提方法可为降低试验成本,提高实际行驶过程排放预测精度提供参考.

关键词： 发动机; 排放; 高斯过程回归; 预测; 外推

本文引用格式

王子垚, 郭凤祥, 陈俐 . 基于外推高斯过程回归方法的发动机排放预测[J]. 上海交通大学学报, 2022 , 56(5) : 604 -610 . DOI: 10.16183/j.cnki.jsjtu.2021.231

Abstract

Aimed at improving the prediction accuracy of engine emissions under driving conditions which are not covered by the training set, an extrapolated Gaussian process regression (GPR) method is proposed. First, the training set data is fed to the GPR model for pre-training, and then a wide-area input set is constructed by uniform sampling between plus/minus three standard deviations, and the hyperparameters are optimized for the goal of minimizing the prediction variance of the input set. The test results on a direct injection gasoline engine show that the mean absolute error of emission prediction using the extrapolated GPR is 0.53411, which is 24.27% lower than that of using the traditional GPR and 36.32% lower than that of using the back propagation (BP) neural network, which signifies the effectiveness of the proposed method in terms of reducing test costs and improving the accuracy of emission prediction during real driving.

Key words： engine; emission; Gaussian process regression (GPR); prediction; extrapolation

参考文献

[1] 张文. 基于运行工况的发动机基本标定研究[D]. 上海: 上海交通大学, 2019.
[1] ZHANG Wen. Research on basic calibration of engine based on operating conditions[D]. Shanghai: Shanghai Jiao Tong University, 2019.
[2] TURKSON R F, YAN F W, ALI M K A, et al. Artificial neural network applications in the calibration of spark-ignition engines: An overview[J]. Engineering Science and Technology, an International Journal, 2016, 19(3): 1346-1359.
[3] YU Q, YANG Y, XIONG X, et al. Assessing the impact of multi-dimensional driving behaviors on link-level emissions based on a portable emission measurement system (PEMS)[J]. Atmospheric Pollution Research, 2021, 12(1): 414-424.
[4] 陈婷, 倪红, 谷雪景, 等. 中国移动源下阶段排放法规综述和分析[J]. 内燃机工程, 2018, 39(6): 24-30.
[4] CHEN Ting, NI Hong, GU Xuejing, et al. An update on China’s mobile source emission regulations[J]. Chinese Internal Combustion Engine Engineering, 2018, 39(6): 24-30.
[5] RAKOPOULOS C D, RAKOPOULOS D C, GIAKOUMIS E G, et al. Validation and sensitivity analysis of a two zone Diesel engine model for combustion and emissions prediction[J]. Energy Conversion and Management, 2004, 45(9/10): 1471-1495.
[6] FINESSO R, SPESSA E. Real-time predictive modeling of combustion and NOx formation in diesel engines under transient conditions[DB/OL]. (2012-04-16)[2021-06-05]. https://www.sae.org/publications/technical-papers/content/2012-01-0899/.
[7] LEFEBVRE A H. Fuel effects on gas turbine combustion-liner temperature, pattern factor, and pollutant emissions[J]. Journal of Aircraft, 1984, 21(11): 887-898.
[8] HANZEVACK E L, LONG T W, ATKINSON C M, et al. Virtual sensors for spark ignition engines using neural networks[C]// Proceedings of the 1997 American Control Conference. Albuquerque, NM, USA: IEEE, 1997: 669-673.
[9] SHI C, JI C W, WANG H Y, et al. Comparative evaluation of intelligent regression algorithms for performance and emissions prediction of a hydrogen-enriched Wankel engine[J]. Fuel, 2021, 290: 120005.
[10] NIU X X, YANG C L, WANG H C, et al. Investigation of ANN and SVM based on limited samples for performance and emissions prediction of a CRDI-assisted marine diesel engine[J]. Applied Thermal Engineering, 2017, 111: 1353-1364.
[11] NALEPA J, KAWULOK M. Selecting training sets for support vector machines: A review[J]. Artificial Intelligence Review, 2019, 52(2): 857-900.
[12] 戴金池, 庞海龙, 俞妍, 等. 基于LSTM神经网络的柴油机NOx排放预测[J]. 内燃机学报, 2020, 38(5): 457-463.
[12] DAI Jinchi, PANG Hailong, YU Yan, et al. Prediction of diesel engine NOx emissions based on long-short term memory neural network[J]. Transactions of CSICE, 2020, 38(5): 457-463.
[13] 秦静, 郑德, 裴毅强, 等. 基于PSO-GPR的发动机性能与排放预测方法[J/OL]. 吉林大学学报(工学版). ( 2021-04-01)[2021-06-05]. https://doi.org/10.13229/j.cnki.jdxbgxb20210111.
[13] QIN Jing, ZHENG De, PEI Yiqiang, et al. Prediction method of engine performance and emission based on PSO-GPR[J/OL]. Journal of Jilin University (Engineering and Technology Edition). (2021-04-01)[2021-06-05]. https://doi.org/10.13229/j.cnki.jdxbgxb20210111.
[14] PUKELSHEIM F. The three sigma rule[J]. The American Statistician, 1994, 48(2): 88-91.
[15] 盛骤, 谢式千, 潘承毅. 概率论与数理统计[M]. 第5版. 北京: 高等教育出版社, 2019.
[15] SHENG Zhou, XIE Shiqian, PAN Chengyi. Probability theory and mathematical statistics[M]. 5th ed. Beijing: Higher Education Press, 2019.
[16] RASMUSSEN C E, WILLIAMS C. Gaussian processes for machine learning[M]. Cambridge, MA, USA: The MIT Press, 2005.
[17] KLENSKE E D, ZEILINGER M N, SCHÖLKOPF B, et al. Gaussian process-based predictive control for periodic error correction[J]. IEEE Transactions on Control Systems Technology, 2016, 24(1): 110-121.
[18] WILSON A G, ADAMS R P. Gaussian process kernels for pattern discovery and extrapolation[C]// International conference on machine learning. Atlanta, USA: PMLR, 2013: 1067-1075.
[19] MOHAMAD M A, SAPSIS T P. Sequential sampling strategy for extreme event statistics in nonlinear dynamical systems[J]. PNAS, 2018, 115(44): 11138-11143.
[20] 中华人民共和国环境保护部, 中华人民共和国国家质量监督检验检疫总局. 轻型汽车污染物排放限值及测量方法(中国第六阶段): GB 18352.6—2016 北京: 中国环境科学出版社. 2020.
[20] Ministry of Environmental Protection of the People’s Republic of China, General Administration of Quality Supervision, Inspection and Quarantine of the People’s Republic of China. Limits and measurement methods for emissions from light-duty vehicles(CHINA 6): GB 18352.6—2016[S]. Beijing: China Environment Science Press, 2020.
[21] STEINIER J, TERMONIA Y, DELTOUR J. Smoothing and differentiation of data by simplified least square procedure[J]. Analytical Chemistry, 1972, 44(11): 1906-1909.
[22] SHLENS J. A tutorial on principal component analysis[EB/OL]. (2014-04-07) [2021-06-05]. https://arxiv.org/abs/1404.1100.
Options
文章导航

/

〈 〉