Knock Detection in Spark Ignition Engines Base on Complementary Ensemble Empirical Mode Decomposition-Hilbert Transform

Abstract

In spark ignition engines, knock onset limits the maximum spark advance. An inaccurate identification of this limit penalises the fuel conversion efficiency. Thus knock feature extraction is the key of closed-loop control of ignition in spark ignition engine. This paper reports an investigation of knock detection in spark ignition (SI) engines using CEEMD-Hilbert transform based on the engine cylinder pressure signals and engine cylinder block vibration signals. Complementary Ensemble Empirical Mode Decomposition (CEEMD) was used to decompose the signal and detect knock characteristic. Hilbert transform was used to analyze the frequency information of knock characteristic. The result shows that, for both of cylinder pressure signals and vibration signals, the CEEMD algorithm could extract the knock characteristic, and the Hilbert transform result shows that the energy of knock impact areas has the phenomenon of frequency concentration in both cylinder pressure signal and cylinder block vibration signal. At last, the knock window is then determined, based on which a new knock intensity evaluation factorKis propose, and it can accurately distinguish between heavy knock, light knock, and normal combustion three states.

The Rossby Normal Mode as a Physical Linkage in a Machine Learning Forecast Model for the SST and SSH of South China Sea Deep Basin

“…The Rossby normal modes are calculated by Equation 5. Since Equation 5 consists of trigonometric functions, Figure 7a is featured by undulating wave train, and it has 5/5, 10/12, 16/18 similarities with field K. The amplitude of field F is different from field K, because the amplitude of signals cannot be well preserved by the band-pass filter, and the ensemble method in EMD algorithm (e.g., CEEMD) can present the characteristics of the original signal accurately (Bi et al, 2016). However, in the perspective of the positive-negative spatial pattern, similarity between field F and field K is very high.…”

Engine block vibrations: An indicator of knocking in the SI engine

“…In-cylinder pressure and the corresponding engine block vibration signals are taken from the research by Fengrong et al [31][32][33]. They had captured the signals from 4 cylinders, 4-stroke, in-line SI engine.…”

“…In this work, the in-cylinder pressure and denoised vibration signals are analyzed to extract the features of knock. The signals are decomposed using EMD, even though there are many improved models like EEMD [9,20], Complementary Ensemble Empirical Mode Decom-position [31], etc. This is due to the better adaptability of EMD in decomposing the signals, i.e., it could naturally deal with the non-stationarities and nonlinearities.…”

A Novel Multiple Feature-Based Engine Knock Detection System using Sparse Bayesian Extreme Learning Machine

“…Empirical mode decomposition (EMD) [31] is self-adaptable and decomposes a signal directly into several IMFs, which are defined as amplitudemodulated-frequency-modulated signals whose number of local extrema and zerocrossings differ at most by one [13]. For the phenomenon that mode mixing occurs repeatedly in EMD, ensemble empirical mode decompositionEMD (EEMD), which is proposed that decreases the chance of undue mode mixing to a certain extent, was proposed [5]. The IMF in EEMD is characterized as the mean of an ensemble of trials whereby a finite-amplitude white noise signal is added to the decomposed data in each trial;, with this approach increases the increase of computational burden since the data size of IMF is equal to that of the raw data.…”