Hits:

Indexed by:会议论文

Date of Publication:2017-01-01

Included Journals:SCIE、EI、CPCI-S、Scopus

Page Number:9537-9542

Key Words:Combustion phase; Residual gas fraction; Multivariate statistical criterion; Cyclic evolution model

Abstract:Combustion phase and residual gas fraction (RGF) are two important factors affecting fuel efficiency and emission, however, which suffer from the cyclic variation of combustion process. To let combustion phase and RGF track their set points and limit the variances, multivariate statistical criterion based controller design is synthesized in this study. For spark ignition gasoline engines, spark advance and intake variable valve timing are considered as inputs, CA50 and RGF are controlled parameters where CA50 represents combustion phase, cyclic evolution model is identified at a steady-state condition, then Hotelling's T-2 test is applied to check whether CA50 or RGF needs adjustment, while the gain of controller is still designed based on discrete linear model. Finally, the proposed controller is validated on a 6-cylinder 3.5L Toyota gasoline engine and the effect of statistical criterion is analyzed in the experiments.