Identifying the release history of a groundwater contaminant source based on an ensemble surrogate model

摘要

In identifying groundwater contaminant sources, given that the simulation model is computationally inefficient, an ensemble surrogate model is proposed to improve the accuracy and robustness of results. The proposed ensemble surrogate model in this paper consists of the following three individual surrogate models: Kriging, radial basis functions and least squares support vector machines. The Adaptive Metropolis-Markov Chain Monte Carlo method is used to assign weights to the three models. Accuracy and robustness of the ensemble surrogate model were tested on not only conservative contaminants but also contaminants containing chemical reaction. The results indicated that the proposed ensemble surrogate model is an effective method to solve the inverse contaminant source identification problems with a high degree of accuracy and short computation time.