LPV-Based Real-Time Intake Oxygen Fraction Estimation of a Turbocharged Diesel Engine with EGR

Abstract

This paper utilizes a Kalman filter observer schemes to estimate the intake manifold oxygen fraction (IOF) of a turbocharged diesel engine with exhaust gas re-circulation (EGR) in real-time. The IOF has a strong correlation with the diesel engine NOx emissions and it is a good feedback candidate for closed-loop NOx control. The dynamics of target diesel engine air-path system is analyzed and used for developing a linear parameter-varying (LPV) model for the observer design. The LPV model based Kalman filter observer is developed and the tuning method for noise covariance matrices is studied to optimize the estimator performance. The estimation performance at steady-state and transient conditions (HD FTP cycle) are validated in a co-simulation environment using GT-SUITE and MATLAB/SIMULINK.