Control of a SCR system using a virtual NOx sensor

This paper evaluates the potential of a model-based control strategy for an automotive Selective Catalytic Reduction (SCR) system. The control strategy is based on the feed-forward / feedback control of NH3 storage and requires the measurement of NOx concentrations upstream and downstream of the catalyst. A robustness analysis of this controller towards upstream NOx measurement or estimation is proposed. The performances of the control strategy with and without an upstream NOx sensor are compared in simulation. The impact of a misevaluation of the upstream NOx concentration on the SCR conversion efficiency is also studied. The accuracy requirements for the air path sensors, compatible with the fulfillment of NOx standards, are given.