Fouling Detection in Heat Exchangers using Extended Kalman Filter

Heat exchangers are important equipment widely used in industries for heat transfer. However, after a long operation, the efficiency of heat transfer may be reduced due to the deposition of impurities on the heat transfer surface, called, fouling. In practice, detection of fouling in heat exchangers is needed for diagnostic, monitoring, and maintenance purposes. This research proposes a fouling detection method using Extended Kalman Filter. Heat exchangers are simulated using cell-based model. Asymptotic fouling model is considered. In detection of fouling in heat exchangers, overall heat transfer coefficient will be estimated using Extended Kalman Filter. The required information in Extended Kalman Filter includes measurement of outlet temperatures on hot and cold sides and heat exchanger models. The results show that for perfect model case, i.e., using same model in simulated heat exchanger and Extended Kalman Filter, overall heat transfer coefficient can be properly estimated. However, for model mismatch case, an offset in estimation can be observed. Although there is an existence of offset in estimation, Extended Kalman Filter can predict the trend of change in heat transfer coefficient quite well.