Estimation of Ambient Air PM_۲.۵ Concentration Using MLP and RBF

Background: Exposure to air pollutants, such as PM۲.۵ is recognized as a significant health risk, contributing to the development of various diseases, and increased risk of premature mortality.Methods: Multilayer perceptron (MLP) and radial basis function (RBF) neural networks, were used to predict the hourly concentration of PM۲.۵ in Isfahan, Iran. The MLP model was designed with five input variables, including PM۲.۵ concentration and weather characteristics, ten hidden layers, and a single output layer. The dataset was divided into three subsets: ۷۰% for training, ۱۵% for testing, and ۱۵% for validation.Results: The results showed that the average concentration of PM۲.۵ was ۲۶.۵ μg/m۳. The root mean square error (RMSE) was estimated as ۶.۴۹ μg/m۳. Increasing the input data resulted in a slight reduction in network error, with the RBF model, utilizing ۱۴۵۰ inputs and an RMSE of ۶.۴۷, achieving the same accuracy as the MLP model with ۱۰ inputs.Conclusion: Given that the PM۲.۵ concentration estimates from the RBF and MLP models deviated by less than ۲۳ and ۲۵%, respectively, compared to the observed concentrations, both MLP and RBF can be regarded as reliable tools for predicting PM۲.۵ levels.

کلیدواژه ها:

Artificial Neural Network ، RBF ، MLP ، Particulate Matter ، Isfahan