Development and Validation of a Knock Prediction Model for Methanol-Fuelled SI Engines

Paper #:
  • 2013-01-1312

Published:
  • 2013-04-08
Citation:
Vancoillie, J., Sileghem, L., and Verhelst, S., "Development and Validation of a Knock Prediction Model for Methanol-Fuelled SI Engines," SAE Technical Paper 2013-01-1312, 2013, https://doi.org/10.4271/2013-01-1312.
Pages:
17
Abstract:
Knock is one of the main factors limiting the efficiency of spark-ignition engines. The introduction of alternative fuels with elevated knock resistance could help to mitigate knock concerns. Alcohols are prime candidate fuels and a model that can accurately predict their autoignition behavior under varying engine operating conditions would be of great value to engine designers.The current work aims to develop such a model for neat methanol. First, an autoignition delay time correlation is developed based on chemical kinetics calculations. Subsequently, this correlation is used in a knock integral model that is implemented in a two-zone engine code. The predictive performance of the resulting model is validated through comparison against experimental measurements on a CFR engine for a range of compression ratios, loads, ignition timings and equivalence ratios.Compared to older correlations that were developed for gasoline, the current autoignition delay correlation captures the high temperature sensitivity of methanol autoignition kinetics. This results in a better prediction of the knock limited spark advance for variations in compression ratio and load. Also the deterioration of knock as a function of spark advance is well reproduced for these conditions.The largest model inaccuracies appear when changing equivalence ratio. Knock tendency is consistently overpredicted for rich mixtures. This is probably due to the effect of evaporation cooling and wall heat transfer which are not well captured by the current model. Further model improvements should therefore focus on these thermal processes inside the cylinder.
Access
Now
SAE MOBILUS Subscriber? You may already have access.
Buy
Select
Price
List
Download
$27.00
Mail
$27.00
Members save up to 40% off list price.
Share
HTML for Linking to Page
Page URL

Related Items

Training / Education
2017-12-18
Technical Paper / Journal Article
2010-10-25
Technical Paper / Journal Article
2010-10-25