Adaptive system of forecasting the air quality

摘要

In the study the adaptive system of forecasting pollutant levels has been proposed. The data for calculations were obtained from an automatic monitoring station located in a Polish city. The system was presented on the example of the suspended particulate matter concentration PM_(10). The proposed method is based on the application of the neural networks namely RBF type which evolves from the data set compiled for the period of a few months. The range of concentrations is divided into five classes of air quality. On the basis of this model one may forecast the level of pollution a few days ahead.