Investigation of Combustion Phasing Control Strategy During Reactivity Controlled Compression Ignition (RCCI) Multicylinder Engine Load Transitions
JOURNAL OF ENGINEERING FOR GAS TURBINES AND POWER-TRANSACTIONS OF THE ASME
Authors: Wu, Yifeng; Hanson, Reed; Reitz, Rolf D.
Abstract
The dual fuel reactivity controlled compression ignition (RCCI) concept has been successfully demonstrated to be a promising, more controllable, high efficiency, and cleaner combustion mode. A multidimensional computational fluid dynamics (CFD) code coupled with detailed chemistry, KIVA-CHEMKIN, was applied to develop a strategy for phasing control during load transitions. Steady-state operating points at 1500 rev/min were calibrated from 0 to 5 bar brake mean effective pressure (BMEP). The load transitions considered in this study included a load-up and a load-down load change transient between 1 bar and 4 bar BMEP at 1500 rev/min. The experimental results showed that during the load transitions, the diesel injection timing responded in two cycles while around five cycles were needed for the diesel common-rail pressure to reach the target value. However, the intake manifold pressure lagged behind the pedal change for about 50 cycles due to the slower response of the turbocharger. The effect of these transients on RCCI engine combustion phasing was studied. The CFD model was first validated against steady-state experimental data at 1 bar and 4 bar BMEP. Then the model was used to develop strategies for phasing control by changing the direct port fuel injection (PFI) amount during load transitions. Specific engine operating cycles during the load transitions (six cycles for the load-up transition and seven cycles for the load-down transition) were selected based on the change of intake manifold pressure to represent the transition processes. Each cycle was studied separately to find the correct PFI to diesel fuel ratio for the desired CA50 (the crank angle at which 50% of total heat release occurs). The simulation results showed that CA50 was delayed by 7 to 15 deg for the load-up transition and advanced by around 5 deg during the load-down transition if the precalibrated steady-state PFI table was used. By decreasing the PFI ratio by 10% to 15% during the load-up transition and increasing the PFI ratio by around 40% during the load-down transition, the CA50 could be controlled at a reasonable value during transitions. The control strategy can be used for closed-loop control during engine transient operating conditions. Combustion and emission results during load transitions are also discussed.
On-board Map Learning-based Combustion Phase Control in Spark Ignition Engines
2017 IEEE CONFERENCE ON CONTROL TECHNOLOGY AND APPLICATIONS (CCTA 2017)
Authors: Zhang, Yahui; Shen, Tielong
Abstract
The combustion phase control is a significant research topic in spark ignition (SI) engines since it affects the fuel efficiency, combustion variability and knocking. Managing the combustion phase, which can be set in the crank angle of 50% mass burnt (CA50), at the optimal reference value where the maximal thermal efficiency can be obtained is challenging due to the engine transient operation, the cyclic variability of combustion process and the optimal operation point drift caused by engine aging, fuel quality, etc. This research addresses this problem in two loops: an outer loop providing the optimal CA50 reference (CA50*) and an inner loop tracking CA50* by managing the spark advance (SA). CA50* is obtained by looking up a 3-dimensional map that can be updated by the stochastic gradient-based on-board map learning algorithm. Then a SA feedforward-feedback controller is designed to track CA50*. The closed loop is of benefit to the fast response of transient operation condition and the on-board learning loop compensates the map drift. Experimental validations of the proposed scheme have been carried out on a six-cylinder SI gasoline engine test bench at transient operation mode.