Use of Kalman Filter and Particle Filter in a One Dimensional Leachate Transport Model

Abstract

Modeling the behavior of pollutants during the flow of leachate through soil is important in predicting the fate of the pollutants. A one dimensional transport model with advection and dispersion is used as the deterministic model of benzene leachate transport from an industrial landfill. A Particle filtering with sequential importance re-sampling (SIR) filter and a classical Kalman filtering were used to predict the benzene plume transport. A traditional root mean square error (RMSE) is used to compare the effectiveness of the Kalman filtering and Particle filtering. The results show that Kalman filter performs better than Particle filter at the initial several time steps. Both Kalman filter and Particle filter can reduce up to 80% error in comparison to a conventional numerical approach.