Measurement of Engine Misfire in the Lamborghini 533 V-12 Engine Using Crankshaft Speed Fluctuations

Active torsional vibration control of reciprocating engines

“…Klenk, Moser, Mueller, & Wimmer, 1993;Azzoni et al, 1995;Williams, 1996), and subsequently for detecting and isolating nonuniform cylinder contributions and estimation of cylinderwise torque contributions (e.g. Kiencke & Nielsen, 2005; Taraza et al, 1998;Moskwa & Wang, 2000;Straky, 1997;Lee et al, 2001).…”

Residual generation and statistical pattern recognition for engine misfire diagnostics

“…Spatial frequency-domain methods rely on Discrete Fourier Transforms (DFT) of measured speed and estimates of the crankshaft frequency response to perform the deconvolution in the frequency domain [2][3][4][5][6]. Another popular approach has been to design an observer which estimates the unknown cylinder torques or pressures based on a model of the crankshaft rotational dynamics [7][8][9][10][11][12][13].…”

Multicylinder engine pressure reconstruction using NARX neural networks and crank kinematics

“…The limited. Recurrent networks are generally much more accurate than feedforward networks but are also FRF approach proposed in reference [12] and simplified in reference [13] suffered from low coherence at much more difficult to train. In this paper, fully recurrent NARX neural networks are adopted for particular speeds, justifying 'averaging' and 'cepstral smoothing'.…”

“…The objectives of the paper are to establish the most suitable NARX configuration and recurrent training engines, showing large errors for all cases. A transfer function approach was tested in reference [15] on a algorithm for crank-based reconstruction, and to confirm the accuracy of the method for multicylinder four-cylinder gasoline engine running at 2400 r/min. Reconstruction errors were around 8 per cent for the pressure recovery under appropriate test conditions.…”