PM2.5 Forecast of Beijing Based on Ensemble Empirical Mode Decomposition and BP Neural Network

doi:10.13209/j.0479-8023.2019.041

Acta Scientiarum Naturalium Universitatis Pekinensis ›› 2019, Vol. 55 ›› Issue (4): 615-625.DOI: 10.13209/j.0479-8023.2019.041

Previous Articles Next Articles

PM_2.5 Forecast of Beijing Based on Ensemble Empirical Mode Decomposition and BP Neural Network

REN Xiaochen^1,4, ZOU Silin¹, TANG Xian², WEI Jun^3,†

1. Department of Atmospheric and Oceanic Sciences, School of Physics, Peking University, Beijing 100871 2. Public Meteorological Service Centre, China Meteorological Administration, Beijing 100081 3. School of Atmospheric Sciences, Sun Yat-sen University, Guangzhou 510275 4. 96813 Troops of PLA, Huangshan 245000

Received:2018-08-08 Revised:2018-09-25 Online:2019-07-20 Published:2019-07-20
Contact: WEI Jun, E-mail: junwei(at)pku.edu.cn

基于集合经验模态分解和BP神经网络的北京市PM_2.5预报研究

任晓晨^1,4, 邹思琳¹, 唐娴², 韦骏^3,†

1. 北京大学物理学院大气与海洋科学系, 北京100871 2. 中国气象局公共气象服务中心, 北京 100081 3. 中山大学大气科学学院, 广州 510275 4. 中国人民解放军96813 部队, 黄山 245000

通讯作者: 韦骏, E-mail: junwei(at)pku.edu.cn
基金资助:
国家自然科学基金(41476008, 41576018)和广西壮族自治区特聘专家专项经费(2018B08)资助

Abstract

Abstract:

A hybrid model with ensemble empirical mode decomposition (EEMD) and BP (Back-Propagation) neural network for next-day forecasting of PM_2.5 concentration in Beijing is developed. The results show that the forecast accuracy of the hybrid model is higher than single BP model. The main error comes from the highest frequency component. The input variables of the hybrid model need to contain information about the output variables. The level of pollutant concentration in the early stage has great influence on the prediction result of the models.

Key words: ensemble empirical mode decomposition (EEMD), BP neural network, PM_2.5 forecast

摘要：

利用集合经验模态分解算法(EEMD)和 BP 神经网络组成的混合模型, 对北京城区PM_2.5浓度值进行短期预报。结果表明, 与单独使用BP神经网络模型相比, EEMD-BP混合模型的预报准确率更高; 混合模型高频部分的预报误差是整体误差的主要方面; 混合模型的输入变量中需包含输出变量的信息; 前期污染物浓度的数值对模型的预报结果有较大的影响。

关键词: 集合经验模态分解算法(EEMD), BP神经网络, PM_2.5预报

REN Xiaochen, ZOU Silin, TANG Xian, WEI Jun. PM_2.5 Forecast of Beijing Based on Ensemble Empirical Mode Decomposition and BP Neural Network[J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2019, 55(4): 615-625.

任晓晨, 邹思琳, 唐娴, 韦骏. 基于集合经验模态分解和BP神经网络的北京市PM_2.5预报研究[J]. 北京大学学报自然科学版, 2019, 55(4): 615-625.

Add to citation manager EndNote|Ris|BibTeX

URL: https://xbna.pku.edu.cn/EN/10.13209/j.0479-8023.2019.041

https://xbna.pku.edu.cn/EN/Y2019/V55/I4/615

[1]	WANG Xinlu, HUANG Ran, ZHANG Wenxian, LÜ Baolei, DU Yunsong, ZHANG Wei, LI Bolan, HU Yongtao. Forecasting Ozone and PM_2.5 Pollution Potentials Using Machine Learning Algorithms: A Case Study in Chengdu [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2021, 57(5): 938-950.
[2]	LI Zequn, WEI Jun. Prediction of PM_2.5 Daily Concentration of Guangzhou Based on Neural Network Algorithms [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2021, 57(4): 645-652.
[3]	WANG Bing,LI Xiaodong. Multi-Scale Fluctuation of European Temperature Revealed by EEMD Analysis [J]. Acta Scientiarum Naturalium Universitatis Pekinensis, 2011, 47(4): 627-635.

PM_2.5 Forecast of Beijing Based on Ensemble Empirical Mode Decomposition and BP Neural Network

基于集合经验模态分解和BP神经网络的北京市PM_2.5预报研究

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 3

Recommended Articles

Metrics

PM2.5 Forecast of Beijing Based on Ensemble Empirical Mode Decomposition and BP Neural Network

基于集合经验模态分解和BP神经网络的北京市PM2.5预报研究

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

References

Related Articles 3

Recommended Articles

Metrics

PM_2.5 Forecast of Beijing Based on Ensemble Empirical Mode Decomposition and BP Neural Network

基于集合经验模态分解和BP神经网络的北京市PM_2.5预报研究