Hourly forecasting on PM 2.5 concentrations using a deep neural network with meteorology inputs.

Autor: Liang Y; School of Energy and Power Engineering, Shandong University, Jinan, 250061, China., Ma J; College of Engineering, Northeastern University, 360 Huntington Avenue, Boston, MA, 02115, USA., Tang C; College of Engineering, Northeastern University, 360 Huntington Avenue, Boston, MA, 02115, USA., Ke N; College of Engineering, Northeastern University, 360 Huntington Avenue, Boston, MA, 02115, USA., Wang D; School of Energy and Power Engineering, Shandong University, Jinan, 250061, China. dwang@sdu.edu.cn.
Jazyk: angličtina
Zdroj: Environmental monitoring and assessment [Environ Monit Assess] 2023 Nov 22; Vol. 195 (12), pp. 1510. Date of Electronic Publication: 2023 Nov 22.
DOI: 10.1007/s10661-023-12081-0
Abstrakt: The PM 2.5 (particulate matter with a diameter of fewer than 2.5 µm) has become a global topic in environmental science. The neural network that based on the non-linear regression algorithm, e.g., deep learning, is now believed to be one of the most facile and advanced approaches in PM 2.5 concentration prediction. In this study, we proposed a PM 2.5 predictor using deep learning as infrastructure and meteorological data as input, for predicting the next hour PM 2.5 concentration in Beijing Aotizhongxin monitor point. We efficiently use the parameter's spatiotemporal correlation by concatenating the dataset with time series. The predicted PM 2.5 concentration was based on meteorology changes over a period. Therefore, the accuracy would increase with the period growing. By extracting the intrinsic features between meteorological and PM 2.5 concentration, a fast and accurate prediction was carried out. The R square score reached maximum of 0.98 and remained an average of 0.9295 in the whole test. The average bias of the model is 9 μg on the validation set and 1 μg on the training set. Moreover, the differences between the predictions and expectations can be further regarded as the estimation for the emission change. Such results can provide scientific advice to supervisory and policy workers.
(© 2023. The Author(s), under exclusive licence to Springer Nature Switzerland AG.)
Databáze: MEDLINE
Externí odkaz: