Prediction of Air Pollution Level Using System Identification Methods

For the prediction of the air pollution level, methods used in system identification can be used. In this case, the system will usually have a number of input parameters (mainly meteorological), and one output parameter, the pollutant concentration. For the pollutant generation and dispersion process, a model is to be developed that satisfies the experimental data (process parameters). This mode will be used to predict the values of some pollution level indicators (pollutant concentration, air quality class, etc.). Unlike the usual systems, in the air pollution case it is very difficult to put the real system in different states by programmed changes in input parameters (meteorological parameters), in order to generate experimental data sets to be used in developing the model.