Research on Air Quality Index Prediction Based on Neural Network: Taking Beijing as an Example
- https://doi.org/10.2991/aebmr.k.191225.124How to use a DOI?
- neural network, air quality index, linear regression, principal component analysis, Bayesian algorithm
With the continuous improvement of the level of industrialization, the air pollution situation in China has become more and more serious. In many places, extreme weather such as haze has appeared, which seriously threatens people’s health. Therefore, it is necessary to establish a scientific and reasonable air quality index prediction model. However, there are significant differences in air quality indices in different quarters, that is, the AQI values are significantly seasonal. Therefore, in order to improve the prediction accuracy, the data of different quarters are distinguished, and models of different quarters are established. In this paper, the principal component analysis method is used to analyze the correlation between API value and various meteorological factors, and the correlation factor is used as the input variable of neural network. The number of neurons in different quarters is determined according to the mean square error, and Bayesian normalization is established. The neural network is a model of the algorithm. Finally, the corresponding model was used to predict the air quality index of the corresponding quarter in 2018 based on the winter, spring, summer and autumn air conditions of Beijing from 2014 to 2017. The results show that the forecasting accuracy of each quarter is 88.27%, 92.28%, 94.04%, and 91.01%, respectively. The prediction accuracy of most studies is 70%~90%, and the prediction accuracy is high, which has certain reference value.
- © 2020, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Cong Zhao AU - Xuemei Li PY - 2020 DA - 2020/01/07 TI - Research on Air Quality Index Prediction Based on Neural Network: Taking Beijing as an Example BT - Proceedings of the 5th International Conference on Economics, Management, Law and Education (EMLE 2019) PB - Atlantis Press SP - 681 EP - 687 SN - 2352-5428 UR - https://doi.org/10.2991/aebmr.k.191225.124 DO - https://doi.org/10.2991/aebmr.k.191225.124 ID - Zhao2020 ER -