Optimization of combustion mode duration for lean gasoline engine with NO_Xstorage-capable passive selective catalytic reduction system

Lean-burn gasoline engines have demonstrated promising potentials to achieve higher fuel efficiency than stoichiometric gasoline engines. However, severe concerns arise in lean NO_xemission control. Three-way catalysts (TWCs), which are broadly applied in stoichiometric gasoline engines, fail to achieve high NO_xconversion efficiency in the presence of excessive oxygen. Emerging passive selective catalytic reduction (SCR) systems with NO_xstorage capability on TWC, offer great potential in NO_xemission reduction for lean-burn gasoline engines at low fuel penalty due to on-board ammonia generation in periodic rich operation. The purpose of this paper is to derive local and global optimization algorithms for optimizing lean and rich operation times in each lean-rich period for lean-burn gasoline engines by considering not only fuel penalty associated with NH3 production but also lean/rich switching frequency. Optimization results demonstrate that both local and optimal optimization strategies result in comparable fuel penalties at the same level of mode-switching frequency. However, the genetic algorithm-based global optimization method requires much higher computational load than the local optimization method and thus is less preferred for real-time applications.