یک رویکرد بهینه سازی مبتنی بر کلینیک مصنوعی چند منظوره برای طراحی شبکه های نظارت بر کیفیت آب در حوضه های رودخانه

Water quality monitoring is important for the management of freshwater resources in river basins. Allocation of monitoring stations is the first step in the design of a water quality network. For this task, planning objectives are identified and a Multi-Objective Artificial Bee Colony-based optimization algorithm is designed and implemented in a Geographic Information System framework. Specifically, the number of stations is minimized in a range of values at the same time that the detection of lower compliance areas, the affected population and the relative importance of the river stretches are maximized. The estimation of pollutant parameters such as Biochemical Oxygen Demand, Faecal Coliform Bacteria or Total Dissolved Solids is performed by using the WorldQual model. This allows to objectively allocate monitoring stations to rivers where no real measurements are available, and thus it is especially relevant to allocate water quality stations for the first time. This approach has been tested on the Great Fish River basin (South Africa), finding networks improving the values of the objective functions between 22.22% and 47.83% with respect to the ones of the current network. Moreover, the solution analysis provides insightful and valuable information to the decision maker.