A GA Based Iterative Model for Identification of Unknown Groundwater Pollution Sources Considering Noisy Data

Abstract

Genetic Algorithms have been applied in solving various complex engineering optimization problems. This chapter presented the application of Genetic Algorithms in identifying unknown groundwater pollution sources of an aquifer. The unknown groundwater pollution sources can be identified by using the inverse optimization model. The inverse optimization model minimizes the difference between the simulated and observed concentration at the observation locations for obtaining the unknown pollution sources. However, the model cannot be setup unless and until the number of pollutions sources are not known. As such, an iterative based methodology is used to obtain the exact number of pollution sources along with their source strength. Further, it is not always possible to accurately measure the concentration data in the field. As such an analysis has been carried out to evaluate the model performance when noisy data is used for the prediction of the sources. The performance of the model is evaluated using an illustrative study area.