Abstract
This paper discusses the application of advanced neural network methods to the development of diagnostics for complex, nonlinear dynamical systems, for which accurate, first-principles models either do no exist or are difficult to derive. We consider two approaches to detect and identify failures in these systems. First, neural networks are trained to act as virtual sensors that emulate the performance of laboratory-quality sensors; this approach provides higher quality diagnostic information than is available directly from production sensors. Second, neural networks are trained to emulate nominal (fault-free) system behavior; model-based fault diagnosis is subsequently achieved by detecting significant deviations between actual and predicted system performance. We present experimental evidence of the viability of both approaches for a difficult automotive diagnostic task.
Preview
Unable to display preview. Download preview PDF.
References
G. Rizzoni, Diagnosis of Individual Cylinder Misfire by Signature Analysis of Crankshaft Speed Fluctuations, SAE No. 890884.
W. B. Ribbens and G. Rizzoni, Onboard Diagnosis of Engine Misfires, SAE No. 901768.
J. V. James, James M. Dosdall and Kenneth A. Marko, Misfire Detection in an Internal Combustion Engine, U.S. Patent 5,044,194. September 3, 1994.
L. A. Feldkamp and G. V. Puskorius, Training Controllers for Robustness: Multistream DEKF, Proceedings of the IEEE Conference on Neural Networks, vol. IV, pp. 2377–2382.
G. V. Puskorius and L. A. Feldkamp, Signal Processing by Dynamic Neural Networks with Application to Automotive Misfire Detection. To appear in Proceedings of the 1996 World Congress on Neural Networks, San Diego.
K. A. Marko, J. V. James, T. M. Feldkamp, G. V. Puskorius and L. A. Feldkamp, Training Recurrent Neural Networks for Classification: Realization of Engine Diagnostics. To appear in Proceedings of the 1996 World Congress on Neural Networks, San Diego.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1996 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Marko, K.A., James, J.V., Feldkamp, T.M., Puskorius, G.V., Feldkamp, L.A. (1996). Signal processing by neural networks to create “virtual” sensors and model-based diagnostics. In: von der Malsburg, C., von Seelen, W., Vorbrüggen, J.C., Sendhoff, B. (eds) Artificial Neural Networks — ICANN 96. ICANN 1996. Lecture Notes in Computer Science, vol 1112. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-61510-5_35
Download citation
DOI: https://doi.org/10.1007/3-540-61510-5_35
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-61510-1
Online ISBN: 978-3-540-68684-2
eBook Packages: Springer Book Archive