Qingchun Guo This email address is being protected from spambots. You need JavaScript enabled to view it.1,2, Zhenfang He This email address is being protected from spambots. You need JavaScript enabled to view it.1,3, Shanshan Li1, Xinzhou Li2,4, Jingjing Meng1, Zhanfang Hou1, Jiazhen Liu1, Yongjin Chen1

1 School of Environment and Planning, Liaocheng University, Liaocheng 252000, China
2 State Key Laboratory of Loess and Quaternary Geology, Institute of Earth Environment, Chinese Academy of Sciences, Xi’an 710061, China
3 State Key Laboratory of Urban and Regional Ecology, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing 100085, China
4 CAS Center for Excellence in Tibetan Plateau Earth Sciences, Beijing 100101, China

Download Citation: ||https://doi.org/10.4209/aaqr.2020.03.0097  

  • Download: PDF

Cite this article:

Guo, Q., He, Z., Li, S., Li, X., Meng, J., Hou, Z., Liu, J. and Chen, Y. (2020) Air Pollution Forecasting Using Artificial and Wavelet Neural Networks with Meteorological Conditions. Aerosol Air Qual. Res. https://doi.org/10.4209/aaqr.2020.03.0097




Air quality forecasting is a significant method of protecting public health by providing an early warning against harmful air pollutants. Correlation analysis was used to distinguish the linear associations between air pollution index (API) and meteorological variables in Xi’an and Lanzhou. Wavelet artificial neural networks (WANNs) and artificial neural networks (ANNs) were employed to identify the non-linear relationships between API and meteorological conditions. Twelve algorithms and nineteen network topologies for the ANN and WANN models were considered to search for an optimal model that is suitable for API forecasting. Among the possible variables for an input structure, the three previous days’ API and 16 meteorological factors were the most effective input variables in Xi’an and Lanzhou. And it can also be predicted with high accuracy from API recorded three days earlier. Correlation coefficient analysis showed that API (t) was best related to API (t + 1) at the two stations. Furthermore, the performances of average temperature, average water vapor pressure, minimum temperature, maximum temperature, API (t - 1), API (t - 2) were better than other variables at the two stations. The simulation results reveal that the WANN and ANN models based on Bayesian regularization algorithm can accurately reproduce the API on both sites. It was showed that the WANN model (R = 0.8846 in Xi’an, R = 0.8906 in Lanzhou) is better than the ANN (R = 0.8037 in Xi’an, R= 0.7742 in Lanzhou) during the forecasting stage. The WANN offers an effective approach to reveal the nonlinear relationship between input variables and output variable of the API, by recognizing the historic patterns between them. The results indicate that WANN is suitable for short-term forecasts of API. And the results may provide a theoretical basis for environmental management department.

Keywords: Air pollution; Wavelet artificial neural network; Meteorological factor; Forecast.

Impact Factor: 2.735

5-Year Impact Factor: 2.827

SCImago Journal & Country Rank

Enter your email below to receive latest published articles in your field.