Abstract

IntroductionDue to the health effects caused by airborne pollutants in urban areas, forecasting of air quality parameters is one of the most important topics of air quality research. During recent years, statistical models based on artificial neural networks (ANNs) have been increasingly applied and evaluated for forecasting of air quality.MethodsThe development of ANN and multiple linear regressions (MLRs) has been applied to short-term prediction of the NO2 and NOx concentrations as a function of meteorological conditions. The optimum structure of ANN was determined by a trial and error method. We used hourly NOx and NO2 concentrations and metrological parameters, automatic monitoring network during October and November 2012 for two monitoring sites (Abrasan and Farmandari sites) in Tabriz, Iran.ResultsDesigning of the network architecture is based on the approximation theory of Kolmogorov, and the structure of ANN with 30 neurons had the best performance. ANN trained by scaled-conjugate-gradient (trainscg) training algorithm has implemented to model. It also demonstrates that MLP neural networks offer several advantages over linear MLR models. The results show that the correlation coefficient (R2) values are 0.92 and 0/94 for NO2 and NOx concentrations, respectively. But in MLR model, R2 values were 0.41 and 0.44 for NO2 and NOx concentrations, respectively.ConclusionsThis work shows that MLP neural networks can accurately model the relationship between local meteorological data and NO2 and NOx concentrations in an urban environment compared to linear models.

Highlights

  • Due to the health effects caused by airborne pollutants in urban areas, forecasting of air quality parameters is one of the most important topics of air quality research

  • It is found that artificial neural networks (ANNs) is a useful tool for the short-term prediction of NO2 and NOx concentrations

  • An ANN trained by scaledconjugate-gradient training algorithm has been implemented to model NO2 and NOx concentrations

Read more

Summary

Methods

The development of ANN and multiple linear regressions (MLRs) has been applied to short-term prediction of the NO2 and NOx concentrations as a function of meteorological conditions. The optimum structure of ANN was determined by a trial and error method. We used hourly NOx and NO2 concentrations and metrological parameters, automatic monitoring network during October and November 2012 for two monitoring sites (Abrasan and Farmandari sites) in Tabriz, Iran

Results
Conclusions
Introduction
Results and discussion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call