Ozone concentration forecasting with neuro-fuzzy approaches

摘要

Forecasting is a challenging problem in highly nonlinear dynamic systems. The main goal in development of forecasting models in complex systems is to produce a model that can accurately behave similar to the main system. In problems such as air pollution forecasting, the presence of uncertainties and nonlinearities affects the model's precision. In this paper, ozone concentration, which is well-known as an index for air pollution is forecasted using neuro-fuzzy models. Causal variables are integrated in the models in order to enhance the model's performance. The results are compared to a fuzzy logic approach to demonstrate reliability and accuracy of the proposed model using real observed data.