Cylinder pressure resonant frequency cyclic estimation-based knock intensity metric in combustion engines. (25th July 2019)

• Record Type: Journal Article
• Authors: Shen, Xun; Zhang, Yahui; Shen, Tielong
• Is Part Of: Applied thermal engineering. Volume 158(2019)
• Publication Date: 2019-07-25
• Subjects: Knock intensity metric -- Pressure oscillation -- Resonant frequency cyclic estimation -- Online calculation -- Internal combustion engines; Heat engineering -- Periodicals; Heating -- Equipment and supplies -- Periodicals; Periodicals; 621.40205
• DOI: 10.1016/j.applthermaleng.2019.113756
• Languages: English
• ISSNs: 1359-4311

Abstract

Highlights: A resonance-based knock intensity metric is proposed for knock quantification. The cyclic pressure resonant frequency is estimated from the trapped mass. The PSD over the estimated resonant frequency is calculated as the metric. The metric calculation method is calibrated and validated with experimental data. Statistical analysis shows that the metric exhibits a high signal-to-noise ratio. Abstract: This paper presents a knock intensity metric calculation method based on the in-cylinder pressure resonant frequency estimation to improve the quantification performance of knock phenomena. The knock intensity metric is defined as the mean energy over a specific frequency band with the estimated resonance frequency as the center position. The resonance frequency is estimated based on the precalibrated Bessel shifting factor and the online trapped mass estimation. Then, setting the estimated resonance frequency as the center point, the power spectral density of the pressure oscillation is calculated using the precalibrated pressure window over the precalibrated resonance frequency bandwidth. The proposed knock intensity metric is experimentally validated by comparison to the metric calculated over the frequency band of 6–25 kHz. The statistical analysis results show that the proposed metric exhibits a higher signal-to-noise ratio and is a more sensitive manner of knock quantification.