Artificial neural network applications in the calibration of spark-ignition engines: An overview

Abstract Emission legislation has become progressively tighter, making the development of new internal combustion engines very challenging. New engine technologies for complying with these regulations introduce an exponential dependency between the number of test combinations required for obtaining optimum results and the time and cost outlays. This makes the calibration task very expensive and virtually impossible to carry out. The potential use of trained neural networks in combination with Design of Experiments (DoE) methods for engine calibration has been a subject of research activities in recent times. This is because artificial neural networks, compared with other data-driven modeling techniques, perform better in satisfying a majority of the modeling requirements for engine calibration including the curse of dimensionality; the use of DoE for obtaining few measurements as practicable, with the aim of reducing engine calibration costs; the required flexibility that allows model parameters to be optimized to avoid overfitting; and the facilitation of automated online optimization during the engine calibration process that eliminates the need for user intervention. The purpose of this review is to give an overview of the various applications of neural networks in the calibration of spark-ignition engines. The identified and discussed applications include system identification for rapid prototyping, virtual sensing, use of neural networks as look-up table surrogates, emerging control strategies and On-Board Diagnostic (OBD) applications. The demerits of neural networks, future possibilities and alternatives were also discussed.

[1]  Jouko Lampinen,et al.  Bayesian approach for neural networks--review and case studies , 2001, Neural Networks.

[2]  T. Marwala,et al.  Fault classification in structures with incomplete measured data using autoassociative neural networks and genetic algorithm , 2006 .

[3]  Rainer Müller,et al.  Approximation and Control of the Engine Torque Using Neural Networks , 2000 .

[4]  Dong-il Dan Cho,et al.  Variable Structure Control Method for Fuel-Injected Systems , 1993 .

[5]  Graham C. Goodwin,et al.  Adaptive filtering prediction and control , 1984 .

[6]  Matthias Schultalbers,et al.  Investigations on a Catalyst Heating Strategy by Variable Valve Train for SI Engines , 2012 .

[7]  Christopher M. Atkinson,et al.  Dynamic Model-Based Calibration Optimization: An Introduction and Application to Diesel Engines , 2005 .

[8]  David Haussler,et al.  Occam's Razor , 1987, Inf. Process. Lett..

[9]  Christopher M. Atkinson,et al.  Virtual Sensing: A Neural Network-based Intelligent Performance and Emissions Prediction System for On-Board Diagnostics and Engine Control , 1998 .

[10]  M.T. Hagan,et al.  Backpropagation through time for a general class of recurrent network , 2001, IJCNN'01. International Joint Conference on Neural Networks. Proceedings (Cat. No.01CH37222).

[11]  Ulrich Lenz,et al.  Transient Air-Fuel Ratio Control Using Artificial Intelligence , 1997 .

[12]  Xiaoou Li,et al.  Some new results on system identification with dynamic neural networks , 2001, IEEE Trans. Neural Networks.

[13]  Minghui Kao,et al.  Algorithm-in-the-Loop with Plant Model Simulation, Reusable Test Suite in Production Codes Verification and Controller Hardware-in-the-Loop Bench Testing , 2010 .

[14]  C. Lee Giles,et al.  What Size Neural Network Gives Optimal Generalization? Convergence Properties of Backpropagation , 1998 .

[15]  David Haussler,et al.  What Size Net Gives Valid Generalization? , 1989, Neural Computation.

[16]  Seinosuke Hara,et al.  A Continuous Variable Valve Event and Lift Control Device (VEL) for Automotive Engines , 2001 .

[17]  M. Jankovic,et al.  Delta air-charge anticipation for mass air flow and electronic throttle control based systems , 2002, Proceedings of the 2002 American Control Conference (IEEE Cat. No.CH37301).

[18]  M. Georgiopoulos,et al.  Feed-forward neural networks , 1994, IEEE Potentials.

[19]  A R Tahavvor,et al.  PREDICTION OF THE TEMPERATURE OF THE HOLE DURING THE DRILLING PROCESS USING ARTIFICIAL NEURAL NETWORKS (RESEARCH NOTE) , 2014 .

[20]  Sven Meyer,et al.  New Calibration Methods and Control Systems with Artificial Neural Networks , 2002 .

[21]  Sun Hai-zhu,et al.  Curve Fitting in Least-Square Method and Its Realization with Matlab , 2005 .

[22]  Jooyoung Park,et al.  Universal Approximation Using Radial-Basis-Function Networks , 1991, Neural Computation.

[23]  Holger R. Maier,et al.  Recent Advances and Future Challenges for Artificial Neural Systems in Geotechnical Engineering Applications , 2009, Adv. Artif. Neural Syst..

[24]  Kenji Suzuki,et al.  Artificial Neural Networks - Industrial and Control Engineering Applications , 2011 .

[25]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[26]  William B. Ribbens.  Understanding Automotive Electronics: An Engineering Perspective , 1982 .

[27]  Bojan Babić,et al.  TOWARDS IMPLEMENTATION AND AUTONOMOUS NAVIGATION OF AN INTELLIGENT AUTOMATED GUIDED VEHICLE IN MATERIAL HANDLING SYSTEMS , 2012 .

[28]  Hong Chen,et al.  Approximation capability to functions of several variables, nonlinear functionals, and operators by radial basis function neural networks , 1993, IEEE Trans. Neural Networks.

[29]  W. Pitts,et al.  A Logical Calculus of the Ideas Immanent in Nervous Activity (1943) , 2021, Ideas That Created the Future.

[30]  Guang-Bin Huang,et al.  Classification ability of single hidden layer feedforward neural networks , 2000, IEEE Trans. Neural Networks Learn. Syst..

[31]  R. Flierl,et al.  The Third Generation of Valvetrains - New Fully Variable Valvetrains for Throttle-Free Load Control , 2000 .

[32]  Ronald J. Williams,et al.  A Learning Algorithm for Continually Running Fully Recurrent Neural Networks , 1989, Neural Computation.

[33]  Zoran Filipi,et al.  Cam-Phasing Optimization Using Artificial Neural Networks as Surrogate Models-Maximizing Torque Output , 2005 .

[34]  Dingli Yu,et al.  Neural network model-based automotive engine air/fuel ratio control and robustness evaluation , 2009, Eng. Appl. Artif. Intell..

[35]  Amir H. Shamekhi,et al.  A new approach in improvement of mean value models for spark ignition engines using neural networks , 2015, Expert Syst. Appl..

[36]  Gregory N. Hullender,et al.  Learning to rank using gradient descent , 2005, ICML.

[37]  Lee A. Feldkamp,et al.  Neurocontrol of nonlinear dynamical systems with Kalman filter trained recurrent networks , 1994, IEEE Trans. Neural Networks.

[38]  Y Lu,et al.  A Sequential Learning Scheme for Function Approximation Using Minimal Radial Basis Function Neural Networks , 1997, Neural Computation.

[39]  E. Oja.  Simplified neuron model as a principal component analyzer , 1982, Journal of mathematical biology.

[40]  Aleš Florian,et al.  An efficient sampling scheme: Updated Latin Hypercube Sampling , 1992 .

[41]  J. Karl Hedrick,et al.  Sliding Mode Fuel-Injection Controller: Its Advantages , 1991 .

[42]  C. Gray,et al.  A Review of Variable Engine Valve Timing , 1988 .

[43]  Yuichi Nakamura,et al.  Approximation of dynamical systems by continuous time recurrent neural networks , 1993, Neural Networks.

[44]  Jacek Czarnigowski,et al.  Hybrid Air/Fuel Ratio Control Using the Adaptive Estimation and Neural Network , 2000 .

[45]  Guillaume Colin,et al.  Neural Model for Real-Time Engine Volumetric Efficiency Estimation , 2013 .

[46]  D. Vogel.  Trading Up: Consumer and Environmental Regulation in a Global Economy , 1997 .

[47]  Ren-Jye Yang,et al.  Moving Least Square Method for Reliability-Based Design Optimization , 2004 .

[48]  Gilead Tadmor,et al.  Reduced-Order Modelling for Flow Control , 2013 .

[49]  M. Majors,et al.  Neural network control of automotive fuel-injection systems , 1994, IEEE Control Systems.

[50]  David Bailey,et al.  How to develop neural-network applications , 1990 .

[51]  Shang-Liang Chen,et al.  Orthogonal least squares learning algorithm for radial basis function networks , 1991, IEEE Trans. Neural Networks.

[52]  H. Hong,et al.  Review and analysis of variable valve timing strategies—eight ways to approach , 2004 .

[53]  Anthony N. Burkitt,et al.  A Review of the Integrate-and-fire Neuron Model: I. Homogeneous Synaptic Input , 2006, Biological Cybernetics.

[54]  David J. Cole,et al.  Modelling nonlinear vehicle dynamics with neural networks , 2010 .

[55]  Jeffrey M. Pfeiffer,et al.  Replacing Volumetric Efficiency Calibration Look-up Tables with Artificial Neural Network-based Algorithm for Variable Valve Actuation , 2010 .

[56]  T. W. Asmus,et al.  Perspectives on Applications of Variable Valve Timing , 1991 .

[57]  O. Axelsson.  A generalized conjugate gradient, least square method , 1987 .

[58]  Yoshua Bengio,et al.  Learning long-term dependencies with gradient descent is difficult , 1994, IEEE Trans. Neural Networks.

[59]  Henry Leung,et al.  Neural data fusion algorithms based on a linearly constrained least square method , 2002, IEEE Trans. Neural Networks.

[60]  Fumihito Arai,et al.  Swing and locomotion control for a two-link brachiation robot , 1994 .

[61]  Ashish Jain,et al.  Design and Development of Variable Valve Actuation (VVA) Mechanism Concept for Multi-Cylinder Engine , 2015 .

[62]  Anil K. Jain,et al.  Artificial Neural Networks: A Tutorial , 1996, Computer.

[63]  Antonio Pietrosanto,et al.  On-line sensor fault detection, isolation, and accommodation in automotive engines , 2002, IMTC/2002. Proceedings of the 19th IEEE Instrumentation and Measurement Technology Conference (IEEE Cat. No.00CH37276).

[64]  Martin A. Riedmiller,et al.  Advanced supervised learning in multi-layer perceptrons — From backpropagation to adaptive learning algorithms , 1994 .

[65]  Chris Bishop,et al.  Improving the Generalization Properties of Radial Basis Function Neural Networks , 1991, Neural Computation.

[66]  Zoran Filipi,et al.  Variable geometry turbine (VGT) strategies for improving diesel engine in-vehicle response: a simulation study , 2004 .

[67]  Xin Yao,et al.  Simultaneous training of negatively correlated neural networks in an ensemble , 1999, IEEE Trans. Syst. Man Cybern. Part B.

[68]  Zoran Filipi,et al.  Using artificial neural networks for representing the air flow rate through a 2.4 liter VVT engine , 2004 .

[69]  Roberto Battiti,et al.  First- and Second-Order Methods for Learning: Between Steepest Descent and Newton's Method , 1992, Neural Computation.

[70]  Benjamin Berger,et al.  Modeling and Optimization for Stationary Base Engine Calibration , 2012 .

[71]  Simon Haykin,et al.  Neural Networks: A Comprehensive Foundation , 1998 .

[72]  George W. Irwin,et al.  Fault Detection in Internal Combustion Engines using a Semi-Physical Neural Network Approach , 2007 .

[73]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[74]  Donald F. Specht,et al.  A general regression neural network , 1991, IEEE Trans. Neural Networks.

[75]  Shivaram Kamat,et al.  Virtual Sensing of SI Engines Using Recurrent Neural Networks , 2006 .

[76]  Elbert Hendricks,et al.  Transient A/F Ratio Errors in Conventional SI Engine Controllers , 1993 .

[77]  J. K. Hedrick,et al.  A Nonlinear Controller Design Method for Fuel-Injected Automotive Engines , 1988 .

[78]  Michael R. Grimes,et al.  A Neural Network Based Methodology for Virtual Sensor Development , 2005 .

[79]  M. D. McKay,et al.  A comparison of three methods for selecting values of input variables in the analysis of output from a computer code , 2000 .

[80]  Eric A. Wan,et al.  Relating Real-Time Backpropagation and Backpropagation-Through-Time: An Application of Flow Graph Interreciprocity , 1994, Neural Computation.

[81]  C. Beltrami,et al.  AFR control in si engine with neural prediction of cylinder air mass , 2003, Proceedings of the 2003 American Control Conference, 2003..

[82]  Dave Richardson,et al.  Sequential DoE Framework for Steady State Model Based Calibration , 2013 .

[83]  Thierry Denoeux,et al.  Principal component analysis of fuzzy data using autoassociative neural networks , 2004, IEEE Transactions on Fuzzy Systems.

[84]  Paul J. Werbos,et al.  Backpropagation Through Time: What It Does and How to Do It , 1990, Proc. IEEE.

[85]  Rudolf Flierl,et al.  Overview of Current Continuously Variable Valve Lift Systems for Four-Stroke Spark-Ignition Engines and the Criteria for their Design Ratings , 2004 .

[86]  Orazio Giustolisi,et al.  A multi-model approach to analysis of environmental phenomena , 2007, Environ. Model. Softw..

[87]  Carlo N. Grimaldi,et al.  OBD Engine Fault Detection Using a Neural Approach , 2001 .

[88]  Ki-Wook Shin,et al.  Automatic Test-Case Generation for Hardware-in-the-Loop Testing of Automotive Body Control Modules , 2013 .

[89]  Antonio Pietrosanto,et al.  On-line sensor fault detection, isolation, and accommodation in automotive engines , 2003, IEEE Trans. Instrum. Meas..

[90]  Jörn Getzlaff,et al.  A Fully Variable Hydraulic Valve Train Concept with Continuous Measuring of the Valve Lift Movement , 2015 .

[91]  A. V. Olgac,et al.  Performance Analysis of Various Activation Functions in Generalized MLP Architectures of Neural Networks , 2011 .

[92]  Kumpati S. Narendra,et al.  Identification and control of dynamical systems using neural networks , 1990, IEEE Trans. Neural Networks.

[93]  Jorge J. Moré,et al.  The Levenberg-Marquardt algo-rithm: Implementation and theory , 1977 .

[94]  Thomas Dresner,et al.  A Review of Variable Valve Timing Benefits and Modes of Operation , 1989 .

[95]  Dennis N. Assanis,et al.  Effect of Exhaust Valve Timing on Gasoline Engine Performance and Hydrocarbon Emissions , 2004 .

[96]  Terrence J. Sejnowski,et al.  The Computational Brain , 1996, Artif. Intell..

[97]  Mark A. Kramer,et al.  Autoassociative neural networks , 1992 .

[98]  Tom Erkkinen.  Fixed-Point ECU Development with Model-Based Design , 2008 .

[99]  O. Nelles.  Nonlinear System Identification: From Classical Approaches to Neural Networks and Fuzzy Models , 2000 .

[100]  S. Liong,et al.  GENERALIZATION FOR MULTILAYER NEURAL NETWORK BAYESIAN REGULARIZATION OR EARLY STOPPING , 2004 .

[101]  Nicolò Cavina,et al.  Virtual GDI Engine as a Tool for Model-Based Calibration , 2012 .

[102]  Richard D. Braatz,et al.  On the "Identification and control of dynamical systems using neural networks" , 1997, IEEE Trans. Neural Networks.

[103]  Shugang Jiang,et al.  Implementation of Model-Based Calibration for a Gasoline Engine , 2012 .

[104]  Carlo N. Grimaldi,et al.  On Board Diagnosis of Internal Combustion Engines: A New Model Definition and Experimental Validation , 1997 .

[105]  Mark Paul Gravesend Guerrier,et al.  The Development of Model Based Methodologies for Gasoline IC Engine Calibration , 2004 .

[106]  Richard J. Beckman,et al.  A Comparison of Three Methods for Selecting Values of Input Variables in the Analysis of Output From a Computer Code , 2000, Technometrics.

[107]  Wolfgang Salber,et al.  Accelerated Powertrain Development Through Model Based Calibration , 2006 .

[108]  D. M. Jones,et al.  Comparison of black-, white-, and grey-box models to predict ultimate tensile strength of high-strength hot rolled coils at the Port Talbot hot strip mill , 2007 .

[109]  C. Lee Giles,et al.  Overfitting and neural networks: conjugate gradient and backpropagation , 2000, Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks. IJCNN 2000. Neural Computing: New Challenges and Perspectives for the New Millennium.

[110]  Terrence J. Sejnowski,et al.  Neural network learning algorithms , 1988 .

[111]  Guang-Bin Huang,et al.  Extreme learning machine: a new learning scheme of feedforward neural networks , 2004, 2004 IEEE International Joint Conference on Neural Networks (IEEE Cat. No.04CH37541).

[112]  Stefan Jakubek,et al.  Regularisation methods for neural network model averaging , 2015, Eng. Appl. Artif. Intell..

[113]  D. Huntington,et al.  Improvements to and limitations of Latin hypercube sampling , 1998 .

[114]  Robert G. Sargent,et al.  Optimization and response surfaces: Gaussian radial basis functions for simulation metamodeling , 2002, WSC '02.

[115]  T. Ahmad,et al.  A Survey of Variable-Valve-Actuation Technology , 1989 .

[116]  Carlo N. Grimaldi,et al.  Prediction of Engine Operational Parameters for On Board Diagnostics Using a Free Model Technology , 1999 .

[117]  Dennis A. Guenther,et al.  A Primer on Building a Hardware in the Loop Simulation and Validation for a 6X4 Tractor Trailer Model , 2014 .

[118]  Georg Wachtmeister,et al.  Development of Dynamic Models for an HCCI Engine with Fully Variable Valve-Train , 2013 .

[119]  Ramesh C. Jain,et al.  A robust backpropagation learning algorithm for function approximation , 1994, IEEE Trans. Neural Networks.

[120]  Rosangela Ballini,et al.  Book reviews: Application of neural networks to adaptive control of nonlinear systems , 2000 .

[121]  Klaus Lamberg,et al.  Hardware-in-the-Loop Testing in the Context of ISO 26262 , 2012 .

[122]  Pj Clarkson,et al.  ENVIRONMENTAL LEGISLATION AS A DRIVER OF DESIGN , 2003 .

[123]  Tom Ma,et al.  Effect of Variable Engine Valve Timing on Fuel Economy , 1988 .

[124]  M. Forster,et al.  Key Concepts in Model Selection: Performance and Generalizability. , 2000, Journal of mathematical psychology.

[125]  Kishan G. Mehrotra,et al.  Elements of artificial neural networks , 1996 .

[126]  Thomas Kruse,et al.  Advanced Statistical System Identification in ECU-Development and Optimization , 2015 .

[127]  Paul Raymund Nicastri,et al.  Automotive vehicle control challenges in the 21st century , 2000 .

[128]  Johan Lennblad,et al.  Neural Network Based Fast-Running Engine Models for Control-Oriented Applications , 2005 .