A Study of Assessment and Prediction of Water Quality Index Using Fuzzy Logic and ANN Models

Various human activities have been the main causes of surface water pollution. The uneven distribution industrial enterprises in territories river basins Ukraine do not always allow real state quality to be assessed. This article has three purposes: (1) modification Ukrainian method for assessing WQI, taking into account level negative impact most dangerous chemical elements, (2) modeling WQI assessment using fuzzy logic and (3) creation an artificial neural network model prediction WQI. used four input variables calculated one output variable (WQI). In final stage study, six ANN models were analyzed, which differed from each other various loss function optimizers activation functions. optimal results shown with softmax Adam’s optimizer (MAPE = 9.6%; R2 0.964). A comparison MAPE indicators created showed that agreement between forecast target data is satisfactory. novelty this study proposal modify methodology Ukraine. At same time, phased joint use mathematical tools such as effectively evaluate predict values, respectively.