Development and experimental validation of a Dual Extended Kalman Filter for three way catalytic converter

A three way catalytic converter (TWC) is responsible for conversion of engine out hazardous pollutants such as carbon monoxide (CO), nitrogen oxides (NOx) and hydrocarbons (HC) into carbon dioxide (CO2), water (H2O) and (N2) nitrogen. TWC conversion efficiency highly depends on the amount of oxygen stored in the converter, or State of Oxygen (SOX). Estimation of SOX during real time operation is key controlling the TWC behaviour maximizing its efficiency. Moreover, the decrease of TWC conversion efficiency over converter lifetime due to mechanical, chemical and thermal factors, is observed through the decrease of a macroscopic quantity, the Oxygen Storage Capacity (OSC). In this paper, a Dual Extended Kalman Filter (d-EKF) to estimate TWC SOX and OSC is presented. The observer is built upon an experimentally validated, physics-based model of the converter developed by the same authors in a previous work. The nonlinear partial differential equation-based TWC model is properly adapted, through the finite discrete method, for real-time estimation within a vehicle engine-control unit (ECU). Experimental results collected from a TWC instrumented with wide-range oxygen sensors show the ability of the dual observer to robustly estimate both SOX and the catalyst age.