Transient Air-Fuel Ratio Estimation in Spark Ignition Engine Using Recurrent Neural Networks

  • Yanhong Zhang
  • Lifeng Xi
  • James Liu
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4693)


Neural networks is very useful in modeling processes for which mathematical modeling is difficult or impossible. In the present work recurrent neural network (RNN) is used for air-fuel ratio (AFR) estimation in Spark Ignition (SI) Engine. AFR estimation is difficult due to the nonlinearity and dynamic behavior in SI engines. Additionally, delays in engine dynamics limit the performance of engine controller. Estimating AFR a few steps in advance can help engine controller to take care of these. RNN is trained using data from engine simulations in MATLAB/SIMULINK environment. Uncorrelated signals were generated for training and validation. It has been shown that recurrent neural network can predict engine simulations with reasonably good accuracy.


Air-fuel ratio air-fuel ratio estimation recurrent neural network 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Alippi, C., Russis, C., Piuri, V.: A Neural-Network Based Control Solution to Air-Fuel Ratio Control for Automotive Fuel-Injection Systems. IEEE Transactions on Systems, Man, and Cybernetics-Part C: Applications and Reviews 2, 259–268 (2003)CrossRefGoogle Scholar
  2. 2.
    Arsie, I., Pianese, C., Sorrentino, M.A.: Procedure to Enhance Identification of Recurrent Neural Networks for Simulating Air-Fuel Ratio Dynamics in SI Engines. Engineering Applications of Artificial Intelligence 19, 65–77 (2006)CrossRefGoogle Scholar
  3. 3.
    Hendricks, E., Sorenson, S.C.: Mean Value Modelling of Spark Ignition Engines. SAE Technical Paper No. 900616 (1990)Google Scholar
  4. 4.
    Hendricks, E., Jensen, M., Kaidantzis, P., Rasmussen, P., Vesterholm, T.: Transient A/F Ratio Errors in Conventional SI Engine Controllers. SAE Technical Paper No. 930856 (1993)Google Scholar
  5. 5.
    Guo-xiu, L., xin, Z.: Transient Air-Fuel Ratio Control in a CNG Engine Using Fuzzy Neural Networks. Journal of Beijing Institute of Technology 1, 100–103 (2005)Google Scholar
  6. 6.
    Copp, D.G, Burnham, K.J., Lockett, F.P.: Model Comparison for Feedforward Air/Fuel Ratio Control. In: IEEE UKACC International Conference on Control’98 (1998)Google Scholar
  7. 7.
    Narendra, S.K., Parthasarathy, K.: Identification and Control of Dynamical Systems Using Neural Networks. IEEE Transactions on Neural Networks 1, 4–27 (1990)CrossRefGoogle Scholar
  8. 8.
    Haykins, S.: Neural Networks: A Comprehensive Foundation. Pearson Education, London (2002)Google Scholar
  9. 9.
    Nørgaard, M., Ravn, O., Poulsen, N.L., Hansen, L.K.: Neural Networks for Modelling and Control of Dynamic Systems. Springer, Heidelberg (2000)CrossRefzbMATHGoogle Scholar
  10. 10.
    Brian, A.A., Jones, V.K., Powell, J.D., et al.: Adaptive Air-Fuel Ratio Control of A Spark—Ignition Engine. SAE 940373 (1994)Google Scholar
  11. 11.
    Masahiro, N., Akira, O., Ahinichi, A.: Model-Based Fuel InJection Control System for SI Engines. SAE 961188 (1996)Google Scholar
  12. 12.
    Cesre, A., Cosimo, D.R, Vincenzo, P.A.: Fine Control of The Air-To-Fuel Ratio with Recurrent Neural Networks. In: IEEE Instrumentation and Measurement Technology Conferenee, St.PauI, Minnesota, USA, IEEE Computer Society Press, Los Alamitos (1988)Google Scholar
  13. 13.
    Frith, A.M, Gent, C.R, Beaumont, A.J.: Adaptive Control Of Gasoline Engine Air-Fuel Ratio Using Artificial Neural Networks. In: IEEE Artificial Neural Networks Conference, IEEE Computer Society Press, Los Alamitos (1995)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2007

Authors and Affiliations

  • Yanhong Zhang
    • 1
  • Lifeng Xi
    • 1
  • James Liu
    • 2
  1. 1.School of Computer Science and Information Technology, Zhejiang Wanli University, Ningbo, Zhejiang 315100P.R. China
  2. 2.BorgWarner Automotive Components (Ningbo) Co.,Ltd, Ningbo, Zhejiang 315104P.R. China

Personalised recommendations