To address the problem of insufficient analysis of continuous changes and randomness in traditional model-based prediction techniques, a gasoline engine air-fuel ratio feedback control method based on Gaussian Process Regression (GPR) intake quantity prediction is proposed. The simulation analysis results show that compared with the real-time feedback control method of the intake air sensor, the lambda average error of the control method proposed in this paper is reduced by 12% and 29% respectively under the transient operating conditions of the two engines, effectively improving the control accuracy of the air-fuel ratio, while also having strong anti-interference ability.
为满足严格的排放法规要求,空燃比的精确控制成为当前车用汽油机控制的关键技术。文献[1]通过安装在节气门前端的空气流量传感器(Mass air flow sensor,MAFS)测量进气量,从而实现空燃比控制。由于近年来越来越多的新技术应用到汽油发动机中,进气歧管内的流动更加复杂,瞬态工况会造成测量值与实际值存在较大误差[2]。因此,进气流量的有效估计是发动机的控制关键。
2000年以后,高斯过程回归(Guassian process regression,GPR)方法在非线性系统建模领域中得到了广泛应用[11]。随着GPR方法的不断发展,其性能得到了更多学者的认可,被广泛应用于各个领域。文献[12]基于GPR模型提出了空天飞行器气路动态建模方法,能够更加快速、准确地预测气路动态。文献[13]基于GPR算法建立了非参数模型,该方法解决了不同环境下复杂多变性的干扰问题。文献[14]利用组合核函数的GPR方法预测了发动机性能。相比RBF以及其他线性回归方法,GPR方法能够在给出预测结果的同时提供置信区间,并且具有较高的非线性系统预测精度。
LiuJ P, XiaX L, WangY, et al. Development of a fast response high accuracy virtual air flow meter for internal combustion engine applications[C]∥2011 Third International Conference on Measuring Technology and Mechatronics Automation,Shanghai, China, 2011: 1037-1041.
[2]
PiquerasP, De La MorenaJ, SanchisE J, et al. Impact of exhaust gas recirculation on gaseous emissions of turbocharged spark-ignition engines[J]. Applied Sciences, 2020, 10(21): No. 7634.
DuYong-bo. Research on control mode l of in-cylinder high pressure water injection of gasoline engine based on prediction of air intake[D]. Chongqing: School of Energy and Power Engineering, Chongqing University, 2022.
[5]
ZhaoY N, ShenT L, JiaoX H, et al. Air-fuel ratio transient control design for gasoline engines based on individual cylinder air charge estimation[C]∥SAE Technical Paper, 2013-01-0102.
SunBo-qi. The study on intake flow estimation method of range-extender engine and experimental verification[D]. Changchun: College of Automotive Engineering, Jilin University, 2020.
[8]
LeiM, JieL, XuY, et al. Intake air mass observer design based on extended Kalman filter for air-fuel ratio control on SI engine[J]. Energies, 2019, 12(18): NO. 3444.
FengYu, JiaoXiao-hong. Transient cylinder air charge estimation with nonlinear Kalman filter for air-fuel ratio control of gasoline engines[J]. Control Theory and Applications, 2015, 32(4): 546-553.
[11]
XieF Q. Gasoline engine intake flow forecast ctudy of chaotic RBF neural network[C]∥Seventh International Conference on Measuring Technology and Mechatronics Automation,Nanchang, China, 2015: 506-511.
[12]
WangY M, ZhangF J, CuiT, et al. Fault diagnosis for manifold absolute pressure sensor (MAP) of diesel engine based on Elman neural network observer[J]. Chinese Journal of Mechanical Engineering, 2016, 29: 386-395.
ZhangWu-long, LiYue-lin, YangDe-zhi, et al. Research on gas intake prediction of gasoline engine under transition condition based on SVR[J]. Vehicle Engine, 2023(1): 44-51.
[15]
SeegerM. Gaussian processes for machine learning[J]. International Journal of Neural Systems, 2004, 14(2): 69-106.
JiTing-wei, ZhaXu, XieFang-fang, et al. Multi-fidelity aerodynamic modeling method of aerospace vehicles based on Gaussian process regression[J]. Journal of Zhejiang University (Engineering Science), 2023, 57(11): 2314-2324.
YangYing-he, WeiHan-di, FanDi-xia, et al. Optimization method of underwater flapping foil propulsion preformance based on Gaussion process regression and deep reinforcement learning[J]. Journal of Shanghai Jiao Tong University, 2023, 188: 1-16.
[20]
ShiX Y, JiangD G, QianW W, et al. Application of the Gaussian process regression method based on a combined kernel function in engine performance prediction[J]. ACS Omega, 2022, 7(45): No.41732.
[21]
ErikssonL, LindellT, LeufvenO, et al. Scalable component-based modeling for optimizing engines with supercharging, E-boost and turbo compound concepts[J]. SAE International Journal of Engines, 2012, 5(2): 579-595.
[22]
ErikssonL, ThomassonA, EkbergK, et al. Look-ahead controls of heavy duty trucks on open roads-six benchmark solutions[J]. Control Engineering Practice, 2019, 83: 45-66.
[23]
WillmottC J. Advantages of the mean absolute error (MAE) over the root mean square error (RMSE) in assessing average model performance[J]. Climate Research, 2005, 30(1): 79-82.