Abstract
The detection of fouling in exhaust gas recirculation (EGR) coolers of diesel engines should be fast and accurate. This would facilitate deciding an effective strategy to combat fouling and to prolong the lifetime of EGR coolers. In the present study, the propensity of soot deposition in a rectangular EGR cooler is modeled using Kalman filters. Noises, coherent feature of many deposition processes which can be resulted from measurement sensors such as thermocouples or incidental deposit flake-off, are also considered in the model. The Kalman filter minimizes the estimation error covariance by considering the measurement and process noise covariance matrices while it can simultaneously handle the noisy data. The results are characterized with measurement process noise covariance. The relation between these two defines the smoothness and shape of the estimated trend of fouling resistance. Comparisons of the experimental data and the resultant model confirmed the usefulness of the applied method for various operating conditions of an EGR cooler prone to particulate deposition of soot particles. The paper proceeds with the impact of such models in monitoring fouling and taking an appropriate mitigation approach in diesel engines.
