Abstract
A special two-level neural network Multiagent system is used in vehicle engine fault diagnosis. Each agent contains a neural network (NN) for its intelligent. The first level is used to acquire the reasons of the fault and the second level is used to classify the fault. There are a few advantages for using two-level NN multiagent system. When new knowledge of faults is acquired, only the first level needs to be retrained. An agent system is added into the second level of the whole system, and the main structure of the network need not be changed. Under the cooperation of the relative agent system, we can diagnose out complex faults and synthetic faults with less resource and better performance.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
Yongji, W., Jian, T.: Neural Network controlling. Mechanics Press, Beijing (1998)
Bradshaw, J.M.: An Introduction to Software Agents. In: Bradshaw, J.M. (ed.) Software Agents, AAAI Press/ The MIT Press (1997)
Finin, T., Labrou, Y., Mayfield, J.: KQML as an agent Communication Language. In: Bradshaw, J. (ed.) Software Agents, MIT Press, Cambridge (1995)
Shuang, C.: The Neural Network tool box and its application based on Matlab. China Science and Technology University Press, Hefei (1998)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2007 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Wu, X., Gao, X., Sharma, D. (2007). A Multiagent Based Vehicle Engine Fault Diagnosis. In: Apolloni, B., Howlett, R.J., Jain, L. (eds) Knowledge-Based Intelligent Information and Engineering Systems. KES 2007. Lecture Notes in Computer Science(), vol 4693. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-74827-4_68
Download citation
DOI: https://doi.org/10.1007/978-3-540-74827-4_68
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-74826-7
Online ISBN: 978-3-540-74827-4
eBook Packages: Computer ScienceComputer Science (R0)