Abstract
This project is sponsored by GenRad, a leading supplier of vehicle diagnostic and service information solutions for complex electrical and electronic systems in the automotive and aerospace industries. The long term goal is to create more sophisticated diagnostic systems that incorporate signal processing of noise/vibration signals and that can provide global diagnostic information (i.e. fuse information derived from multiple sources).The aim of this project is to test the feasibility of using neural networks and belief nets for condition monitoring and fault diagnosis in the automotive industry. Neural networks are used for signal processing at the individual sensor level, and Bayesian belief nets are used for reasoning about the predictions made by individual neural networks.
| Date of Award | 1999 |
|---|---|
| Original language | English |
| Awarding Institution |
|
Keywords
- diagnostic conditioning monitoring
- information engineering
- automatic