Regeneration Control for a Bypass-Regeneration Lean NOx Trap System

Lean NO_x traps (LNTs) are one of two aftertreatment technologies on the forefront for lean burn engine NO_x remediation, the other technology being urea-SCR. A LNT is capable of surface storage of NO_x from the exhaust during lean engine operation with a high initial trapping efficiency. As the catalyst surface becomes occupied with stored NO_x, the storage efficiency begins declining to nearly zero. At some point during the storage phase, it is necessary to regenerate the catalyst by introducing a supply of rich gasses to chemically reduce the stored surface species, which include both NO_x and oxygen. Regeneration is required on the order of minutes while the regeneration lasts only for a few seconds. This work presents experimental and simulated results of a control algorithm that utilizes catalyst temperature measurements rather than the conventional exhaust gas sensors for a diesel LNT system. Based on a nonlinear, experimentally validated model, a control algorithm is developed based on a linearized model using realistic assumptions regarding the physical system. The control algorithm uses a single temperature measurement inside the catalyst to overcome the issues of observability that are often encountered with storage catalysts