The paper deals with a diagnostic method that allows to detect engine knock. The developed algorithm differentiates three kinds of engine cycles: absence of knock, increasing knock and heavy knock. The decision is taken from a block vibration signal. The diagnostic method is based on pattern recognition. Three models of different data shapes provided from the accelerometer are elaborated. This is done using a time-scale analysis tool called a wavelet network. It allows to extract relevant features from the signal. The aim of the method is then to partition the feature space into classes representing the knock states. Experimental results are reported.
Engine knock detection from vibration signals using pattern recognition
Erkennen des Motorklopfens mittels Vibrationssignalen und einer Bildmustererkennung
1997
9 Seiten, 8 Bilder, 3 Tabellen, 9 Quellen
Conference paper
English
96A4061 Engine knock detection using pattern recognition
British Library Conference Proceedings | 1996
|SAE Technical Papers | 1998
|Automotive engineering | 1998
|Valve event detection using knock sensor signals
Automotive engineering | 2007
|Valve Event Detection Using Knock Sensor Signals
SAE Technical Papers | 2007
|