The quality of a given water body can be assessed through the analysis of a number of indicators. Mathematical and computational models can be built to simulate the behavior of these indicators (observable variables), in such a way that different scenarios can be generated, supporting decisions regarding water resources management. In this study, the transport of a conservative contaminant in an estuarine environment is simulated in order to identify the position and intensity of the contaminant source. For this, it was formulated an inverse problem, which was solved through computational intelligence methods. This approach required adaptations to these methods, which had to be modified to relate the source position to the discrete mesh points of the domain. In this context, two adaptive techniques were developed. In one, the estimated points are projected to the grid points, and in the other, points are randomly selected in the iterative search spaces of the methods. The results showed that the methodology here developed has a strong potential in water bodies’ management and simulation.
An iterative strategy for contaminant source localisation using GLMA optimization and Data Worth on two synthetic 2D Aquifers
