logo

SCIENCE CHINA Information Sciences, Volume 61 , Issue 7 : 070205(2018) https://doi.org/10.1007/s11432-017-9381-6

Optimal calibration scheme for map-based control of diesel engines

More info
  • ReceivedDec 4, 2017
  • AcceptedFeb 19, 2018
  • PublishedJun 12, 2018

Abstract

Map-based control has the advantage of a simple control structure. However, in the case of a complex plant, calibrating the map is difficult. Calibration methodology using plant model is increasing in engine calibration industries. However, the models are for steady-state operation and cannot simulate at transient operation well. This study introduces the dynamic empirical model and applies the transient corrective function of an electronic control unit to simulate more realistic behavior. The optimization problem for calibrating maps of diesel engines is constructed, and the formulation of cost function and constraints is discussed. Consequently, the proposed calibration scheme can find an optimal map in satisfied constraints. Finally, the optimized maps are validated using mass production diesel engines.


Supplement

Appendix

Nomenclature


References

[1] Angridge S, Fessler H. Strategies for High EGR Rates in a Diesel Engine. SAE Technical Paper 2002-01-0961. 2002. Google Scholar

[2] Nishio Y, Hasegawa M, Tsutsumi K, et al. Model Based Control for Dual EGR System with Intake Throttle in New Generation 1.6L Diesel Engine. SAE Technical Paper 2013-24-0133. 2013. Google Scholar

[3] Neely G, Sasaki S, Huang Y, et al. New Diesel Emission Control Strategy to Meet US Tier 2 Emissions Regulations. SAE Technical Paper 2005-01-1091. 2005. Google Scholar

[4] Ohata A. A desired modeling environment for automotive powertrain controls. In: Identification for Automotive Systems. London: Springer, 2012. 13--34. Google Scholar

[5] Baumann W, Dreher T, Röpke K, et al. DoE for series production calibration. In: Proceedings of the 7th Conference on Design of Experiments (DoE) in Engine Development, Berlin, 2013. Google Scholar

[6] Haukap C, Barzantny B, Röpke K. Model-based calibration with data-driven simulation models for non-DoE experts. In: Proceedings of the 6th Conference on Simulation and Testing for Automotive Electronics, Berlin, 2014. Google Scholar

[7] Murata Y, Kato Y, Kanda T, et al. Application of model based calibration to mass production diesel engine development for indian market. In: Proceedings of the 8th Conference on Design of Experiments (DoE) in Engine Development, Berlin, 2015. Google Scholar

[8] Brahma I, Rutland C. Optimization of Diesel Engine Operating Parameters Using Neural Networks. SAE Technical Paper 2003-01-3228. 2003. Google Scholar

[9] Brahma I, Sharp M C, Frazier T R. Empirical modeling of transient emissions and transient response for transient optimization. SAE Int J Engines, 2009, 2: 1433-1443 CrossRef Google Scholar

[10] Berger B, Rauscher F, Lohmann B. Analysing Gaussian processes for stationary black-box combustion engine modelling. IFAC Proc Volumes, 2011, 44: 10633-10640 CrossRef Google Scholar

[11] Mrosek M, Sequenz H, Isermann R. Control oriented NO$_x$ and soot models for diesel engines. IFAC Proc Vol, 2010, 43: 234--239. Google Scholar

[12] Atkinson C, Mott G. Dynamic Model-Based Calibration Optimization: an Introduction and Application to Diesel Engines. SAE Technical Paper 2005-01-0026. 2005. Google Scholar

[13] Atkinson C, Allain M, Zhang H. Using Model-Based Rapid Transient Calibration to Reduce Fuel Consumption and Emissions in Diesel Engines. SAE Technical Paper 2008-01-1365. 2008. Google Scholar

[14] Sakushima N, Wolf B, Karsten R, et al. Transient modeling of diesel engine emissions. Int J Autom Eng, 2013, 4: 63--68. Google Scholar

[15] Berger B. Modeling and optimization for stationary base engine calibration. Dissertation for Ph.D. Degree. Munich: The Technical University of Munich, 2012. Google Scholar

[16] Niedernolte H, Kloepper F, Mitterer A, et al. Workflow for data evaluation during basic calibration of combustion engines. In: Proceedings of IEEE Conference on Computer Aided Control System Design, IEEE International Conference on Control Applications, IEEE International Symposium on Intelligent Control, Munich, 2006. 2060--2065. Google Scholar

[17] Fukuhara K, Murata Y, Nishio Y, et al. Dynamic MBC methodology for transient engine combustion optimization. In: Proceedings of International Calibration Conference Automotive Data Analytics, Methods, DoE, Berlin, 2017. Google Scholar

[18] Shishido T, He J, Kaihatsu M, et al. Dynamic modeling for gasoline direct injection engines. In: Proceedings of International Calibration Conference Automotive Data Analytics, Methods, DoE, Berlin, 2017. Google Scholar

[19] Bishop C M. Pattern Recognition and Machine Learning (Information Science and Statistics). New York: Springer, 2007. 138--147. Google Scholar

[20] Huber T, Hanselmann M, Kruse T. Use of data based models to predict any RDE cycles — challenges, expriences and results. Emission Control Conference 2016, Technische Universit??t Dresden. Google Scholar

[21] Behr L, Zimmermann U, Trinkert S, et al. Increased efficiency in the calibration process of automotive Li-ion battery systems. In: Proceedings of Internationales Stuttgarter Symposium, Wiesbaden, 2016. 101--115. Google Scholar

[22] Baumann W, Schaum S, Ropke K, et al. Excitation signals for nonlinear dynamic modeling of combustion engines. In: Proceedings of the 17th World Congress the International Federation of Automatic Control, Seoul, 2008. Google Scholar

  • Figure 1

    (Color online) Diesel engine system [2]@Copyright 2013 SAE International.

  • Figure 2

    (Color online) Map-based diesel engine control system.

  • Figure 3

    (Color online) Grid index $j$ numbering of $~N~\times~M$ map. (a) Grid index point of basic map; (b) grid index $j$ of basic map.

  • Figure 4

    NARX structure using a Gaussian process model [20]@Copyright 2016 Springer Fachmedien Wiesbaden. protectłinebreak (a) One-step ahead prediction; (b) multi-step ahead prediction.

  • Figure 5

    (Color online) Test plan and training data for plant modeling. (a) Chirp type; (b) APRBS type.

  • Figure 6

    (Color online) Boundary finder [7].

  • Figure 7

    (Color online) Validation results. (a) Measured and predicted of NO$_x$ emission; (b) measured and predicted of soot emission; (c) measured and predicted of CO emission; (d) measured and predicted of fuel mass flow; (e) engine speed.

  • Figure 8

    (Color online) Transient corrective function behavior of ECU. (a) Engine speed; (b) measured and simulated HP-EGR valve position; (c) engine torque; (d) measured and simulated LP-EGR valve position; (e) desired fresh air mass flow; (f) measured and simulated throttle valve position; (g) desired boost pressure; (h) measured and simulated VGT position.

  • Figure 9

    (Color online) Optimized map. (a) Main injection timing; (b) injection pressure; (c) boost pressure; (d) air mass flow.

  • Figure 10

    (Color online) Experiment and simulation of optimized map (NEDC). (a) Measured, predicted and error of NO$_x$ emission; (b) measured, predicted and error of soot emission; (c) measured, predicted and error of CO emission; protectłinebreak (d) measured, predicted and error of fuel mass flow; (e) vehicle speed.

  • Figure 11

    (Color online) Experiment and simulation of optimized map (WLTC). (a) Measured, predicted and error of NO$_x$ emission; (b) measured, predicted and error of soot emission; (c) measured, predicted and error of CO emission; protectłinebreak (d) measured, predicted and error of fuel mass flow; (e) vehicle speed.

  • Table A1   Nomenclature
    Item Description
    APRBS Amplitude modulated pseudo random binary sequences
    BTDC Before top dead center
    DPF Diesel particulate filter
    DOHC Double overhead camshaft
    EAT Exhaust gas after treatment system
    EGR Exhaust gas recirculation
    ECU Electric control unit
    GPR Gaussian process regression
    HP-EGR High pressure loop exhaust gas recirculation
    ICE Internal comubustion engines
    $k$ Discrete time step index
    LP-EGR Low pressure loop exhaust gas recirculation
    MBC Model based calibration
    $N$ Row length of map
    $n$ Row index of map
    NARX Non-linear autoregressive exogenous model
    NEDC New European Driving Cycle
    $M$ Column length of map
    $m$ Column index of map
    RDE Real driving emissions
    $T_e$ Engine torque (Nm)
    $T_{\rm~NEDC}$ discrete time index at the last of NEDC
    $T_{\rm~WLTC}$ discrete time index at the last of WLTC
    $u$ Design variable
    VGT Variable geometry turbocharger
    WLTC worldwide harmonized light vehicles test cycles
    $\omega$ Engine speed (rpm)
    $\delta~t$ Discrete sampling time (s)
    $\rho$ Weighting factor
  • Table 1   nRMSE of validation results
    Item rm nRMSE (%)
    NO$_x$ 2.75
    Soot 4.81
    CO 3.84
    Fuel 1.61
  • Table 2   Specifications of diesel engine
    Item Specification
    EngineInline 4-cylinder DOHC
    Bore $\times$ stroke $\phi$ 76 $\times$ 88 mm
    Displacement 1597 ${\rm~cm}^2$
    Compression ratio16.0
    Max. torque300 Nm / 2000 rpm
    Max. power88 kW / 4000 rpm
    Common-rail fuel injectorSolenoid (Max.180 MPa)
    TurbochargerSingle VGT
    EGR systemHP hot EGR and/or LP cooled EGR

Copyright 2020 Science China Press Co., Ltd. 《中国科学》杂志社有限责任公司 版权所有

京ICP备17057255号       京公网安备11010102003388号