Signal processing pattern classification techniques to improve knock detection in spark ignition engines
✍ Scribed by F. Molinaro; F. Castanié
- Publisher
- Elsevier Science
- Year
- 1995
- Tongue
- English
- Weight
- 525 KB
- Volume
- 9
- Category
- Article
- ISSN
- 0888-3270
No coin nor oath required. For personal study only.
✦ Synopsis
The aim of this paper is to illustrate the contribution of signal processing pattern recognition techniques to the resolution of a physical problem: entire knock detection. Knock is an abnormal combustion of spark ignition in automotive engines. Classical detectors use the energy of engine vibration in order to detect it. Unfortunately, detection becomes erroneous at high speeds when noise level increases. Pattern recognition techniques allow the determination of optimal parameters and methods for knock recognition: cepstral coefficients and amplitude histograms improve knock recognition at 5500 rpm and for other speeds. A comparison between five speeds using the conventional detector and the new detector shows that the latter clearly outperforms the former. The method used here of knock detection can be used in several other applications e.g. gear default detection and axle fissure.