Developing a neural network and real genetic algorithm combined tool for an engine test bed

Abstract

In the automotive industry, engine test engineers are required to deal with a huge quantity of experimental data obtained from engine test beds each day. Those data must be analysed to evaluate engine performance and to guide further engine test operations. In order to improve efficiency and reduce expenditure of time in engine testing, it is very important for engine test bed controllers to develop a mathematical model from existing engine test data. This paper presents an investigation of a neural network-genetic algorithm (GA) combined tool for engine modelling. In the modelling tool, a real-coded GA has been employed to train three different groups of neural networks (a multilayer perceptron group, a radial basis function group, and a bar function networks group) and then finally to find the most suitable neural network model for engine modelling. The experimental results given in this paper show that the proposed tool has been successfully used for Rover engine testing.

Keywords

evolutionary neural networks real-coded genetic algorithm engine modelling engine calibration

Get full access to this article

View all access options for this article.

References

Scaife

Charlton

Mobley

A neural network for fault recognition. SAE paper 930861, 1993.

Stevens

S. P.

Shayler

P. J.

T. H.

Experimental data processing techniques to map the performance of a spark ignition engine. Proc. Instn Mech. Engrs, Part D: J. Automobile Engineering, 1995, 209 (D4), 297–306.

Shayler

P. J.

Darnton

T. H.

Predicting the fuel consumption of vehicles for drive cycles starting from cold ambient conditions. In Proceedings of the 5th International Congress on

Powertrain and Environment, Strasbourg, 1995, paper SIA9506A27.

Brace

C. J.

Deacon

Vaughan

N. D.

Horrocks

R. W.

Burrows

C. R.

An operating point optimizer for the design and calibration of an integrated diesel/continuously variable transmission powertrain. Proc. Instn Mech. Engrs, Part D: J. Automobile Engineering, 1999, 213 (D3), 215–226.

Jocob

Ball

Non-parametric models in the monitoring of engine performance and condition. Part 1: Modelling of non-linear engine processes. Proc. Instn Mech. Engrs, Part D: J. Automobile Engineering, 1999, 213 (D1), 73–81.

Jacob

Ball

Non-parametric models in the monitoring of engine performance and condition. Part 2: Non-intrusive estimation of diesel engine cylinder pressure and its use in fault detection. Proc. Instn Mech. Engrs, Part D: J. Automobile Engineering, 1999, 213 (D2), 135–143.

Zhang

Shi

Reconstructing cylinder pressure from vibration signals based on radial basis function networks. Proc. Instn Mech. Engrs, Part D: J. Automobile Engineering, 2001, 215 (D6), 761–767.

Manzie

Palaniswami

Watson

Gaussian networks for fuel injection control. Proc. Instn Mech. Engrs, Part D: J. Automobile Engineering, 2001, 215 (D10), 1053–1068.

Lin

M. H.

Duan

Engine test data modelling by evolutionary radial basis function networks. Proc. Instn Mech. Engrs, Part D: J. Automobile Engineering, 2003, 217 (D5), 489–497.

10.

Hornick

Stinchcombe

White

Multilayer feedforward networks are universal approximators. Neural Networks, 1989, 2, 359–366.

11.

Hartman

E. J.

Keeler

J. D.

Kowalski

J. M.

Layered neural networks with Gaussian hidden units as universal approximations. Neural Comput., 1990, 2 (2), 210–215.

12.

Bishop

C. M.

Neural networks for pattern recognition, 1995 (Oxford University Press, Oxford).

13.

Hassoun

M. H.

Fundamentals of artificial neural networks, 1995 (MIT Press, Cambridge, Massachusetts).

14.

Nabney

I. T.

NETLAB: Algorithms for pattern recognition, 2001, (Springer-Verlag, UK).

15.

Nørgaard

Ravn

Poulsen

N. K.

Hansen

L. K.

Neural networks for modelling and control of dynamic systems, a practitioner's handbook, 2000 (Springer-Verlag, UK).

16.

Hartman

E. J.

Keeler

J. D.

Predicting the future: Advantages of semilocal units. Neural Comput., 1991, 3 (4), 566–578.

17.

Duch

Jankowski

New neural transfer functions. J. Appl. Math. Comput. Sci., 1997, 7 (3), 639–658.

18.

Goldberg

D. E.

Genetic algorithms in search, optimisation & machine learning, 1989 (Addison-Wesley, Reading, Massachusetts).

19.

Press

W. H.

Flannery

Teukolsky

Vettering

Numerical recipes, 1986 (Cambridge University Press, New York).

20.

Michalewicz

Genetic algorithms + data structures = evolution programs, 3rd edition, 1996 (Springer-Verlag, Berlin).

21.

Herrera

Lozano

Verdegay

J. L.

Tackling real-coded genetic algorithms: Operators and tools for behavioural analysis. Artif. Intell. Rev., 1998, 12 (4), 265–319.

22.

Mühenbein

Schlierkemp-Voosen

Predictive models for the breeder genetic algorithm. I. continuous parameter optimization. Evolutionary Comput., 1993, 1, 25–49.

23.

Eshelman

L. J.

Schaffer

J. D.

Real-coded genetic algorithms and interval schemata. In Foundation of genetic algorithms (Ed. Darrel Whitley

), 1993, pp. 187–202 (Morgan Kaufman).

24.

Prechelt

PROBEN1 — a set of neural network benchmark problems and benchmarking rules, Technical report 21/94, University of Karlsruhe, Karlsruhe, Germany, 1994.

25.

Chen

Billings

S. A.

Grant

P. M.

Non-linear system identification using neural networks. Int. J. Control, 1990, 51 (6), 1191–1214.

26.

Billings

S. A.

Zhu

Q. M.

Non-linear model validation using correlation tests. Int. J. Control, 1994, 60 (6), 1107–1120.

27.

Billings

S. A.

Zhu

Q. M.

Model validation tests for multivariable nonlinear models. Int. J. Control, 1995, 62, 749–766.

28.

Neural Network Toolbox user's guide, 2001 (Math-Works).

29.

Nabney

I. T.

NETLAB Neural Network Toolbox, 2001 (Neural Computing Research Group, Aston University).

30.

Lin

Modelling and optimisation of engine calibration by neural networks and evolutionary computation. PhD Thesis, University of Derby, 2005.