Abstract: Knock feature extraction is the key of closed loop control of ignition in gasoline engine. Based on the ensemble empirical mode decomposition (EEMD), a gasoline engine knock feature extraction method is presented. By adding some finite amplitude Gaussian white noises to the signal, EEMD keeps the signal continuous in different time span, and therefore the mode mixing inhering in the classical empirical mode decomposition (EMD) method is alleviated. The effectiveness of applying EEMD and EMD to extract knock feature from a pressure signal measured from combustion chamber and a vibration signal measured from cylinder head are investigated. The experimental results demonstrate that the knock feature can be effectively extracted from both pressure and vibration signals by EEMD. On the other hand, owing to the interference in different modes, the knock signature can only be extracted from the pressure signal by EMD.

Key words: empirical mode decomposition, feature extraction, gasoline engine, knock