Combustion Variation Control of SI Engines via Hypothesis Testing and EGR Step Valve. Issue 10 (2021)

• Record Type: Journal Article
• Authors: Zhao, Kai; Shi, Haoyun; Shen, Tielong
• Is Part Of: IFAC-PapersOnLine. Volume 54:Issue 10(2021)
• Page Start: 96
• Page End: 101
• Publication Date: 2021
• Subjects: Engine -- Combustion variation -- Hypothesis testing -- EGR; Automatic control -- Periodicals; 629.805
• DOI: 10.1016/j.ifacol.2021.10.147
• Languages: English
• ISSNs: 2405-8963

Abstract

Abstract: In gasoline engines, the variation of the combustion is often evaluated by the coefficient of variance (CoV) of the indicated mean effective pressure (IMEP). Due to the involvement of the mean and the standard deviation in the calculation, a large number of engine cycles are necessary for an accurate estimation of the CoV. However, in the case of realtime engine control, the number of available data is often limited. Thus, statistic-based methods have to be applied to determine whether the CoV is greater or less than a given threshold. In this work, a hypothesis testing based on McKay's approximation is proposed to statistically evaluate the CoV of IMEP. The results of the proposed testing method are then used to control the EGR valve step to maintain the IMEP CoV at a given threshold. The proposed method was validated on a testbench with a production gasoline engine.