Search Results

This paper presents a model-based control system for SCR urea dosing employing an embedded real-time SCR chemistry model and a NH3 sensor. The control algorithm consists of a number of control features designed to enhance ammonia storage control and closed-loop compensation using the mid-brick NH3 sensor. An adaptive control algorithm is developed to demonstrate robustness of the feedback control system to compensate for catalyst aging, urea injection malfunction, or dosing fluid concentration variation. Simulation and engine dynamometer testing following ESC and FTP emission cycles are used to demonstrate the advantages of this control approach for meeting both NOx emission requirements and NH3 slip targets. Furthermore this paper demonstrates potential of a NH3 sensor for on-board diagnostics. Additionally the feasibility of implementing model based algorithms in a 32-bit floating-point environment with an automotive controller is examined.

The Cu-zeolite (CuZ) SCR catalyst enables higher NOx conversion efficiency in part because it can store a significant amount of NH3. “NH3 storage control”, where diesel exhaust fluid (DEF) is dosed in accord with a target NH3 loading, is widely used with CuZ catalysts to achieve very high efficiency. The NH3 loading actually achieved on the catalyst is currently estimated through a stoichiometric calculation. With future high-capacity CuZ catalyst designs, it is likely that the accuracy of this NH3 loading estimate will become limiting for NOx conversion efficiency. Therefore, a direct measurement of NH3 loading is needed; RF sensing enables this. Relative to RF sensing of soot in a DPF (which is in commercial production), RF sensing of NH3 adsorbed on CuZ is more challenging. Therefore, more attention must be paid to the “microwave resonance cavity” created within the SCR assembly. The objective of this study was to develop design guidelines to enable and enhance RF sensing.