Generating Automotive Electrical System Models from Component Based Qualitative Simulation

Type Conference Proceeding (Non-Journal item)
Original languageEnglish
Title of host publicationApplications and Innovations in Intelligent Systems IX
Subtitle of host publicationProceedings of ES2001, the Twenty-first SGES International Conference on Knowledge Based Systems and Applied Artificial Intelligence, Cambridge, December 2001
EditorsAnn Macintosh, Mike Moulton, Alun Preece
PublisherSpringer Nature
Pages100-114
Number of pages15
ISBN (Electronic)978-1-4471-0149-9
ISBN (Print)978-1-85233-530-4
Publication statusPublished - 11 Dec 2001
Links
View graph of relations
Citation formats

Abstract

Model based reasoning applied to electrical systems has matured over recent years resulting in deployment of commercial design analysis tools in the automotive industry [1, 41. These tools work at the component level on individual systems or subsystems. Analysis of multiple systems is becoming necessary because of the increase in system and subsystem interactions resulting from Electronic Control Unit based system architectures.

To address the resulting simulation complexity this paper considers the automated construction of black box system or subsystem models from envisionments produced by qualitative simulation. The models can then be used to improve efficiency of the repetitive simulation procedures required to perform automated Failure Mode Effects Analysis (FMEA).

The approach also has the potential to allow Sneak Circuit Analysis (SCA)[2, 31 to be extended to include systems with internal memory. It provides the basis of a useful visualization tool for complex system behaviour.