基于lstm算法和wrf-...空气质量变化影响的定量分析 亓浩雲.pdf

: .
中国环境科学度均有提升且 2021 疫情时期颗粒物浓度同样有提高,2020 疫情时期其他污染物浓度均有不同程度的改善,而与 2019 历史同期相比,两次疫情期
间臭氧浓度同样有升高现象,除此以外,2021 LSTM(Long Short-Term Memory)算法和 WRF(Weather Research And
Forecasting Model)-CMAQ(Community Multiscale Air Quality)模型量化了两次疫情时期气象因素对于污染物浓度变化的影响,根据空气质量模拟法
,LSTM 算法在两次疫情期间的模拟均显示人为影响对污染物产生了负影响（降低了污染
物浓度）且在总变化影响中占比较高,而 CMAQ 模式模拟结果中的气象因素影响占比远高于 LSTM 模式在两次疫情模拟中表现出了
不同的结果,在 2020 疫情中人为影响占据了主导,而在 2021 疫情中,相比较 2020 疫情时期,除 NO2 外,人类活动对其他污染物的影响均为正值（促进
了污染物浓度升高）.
关键词:机器学习算法;空气质量模型;新冠肺炎;邢台市
中图分类号:X511 文献标识码:A

Quantitative analysis of the impact of different factors on air quality changes in Xingtai City during the epidemic period
based on LSTM algorithm and WRF-CMAQ model. QI Hao-yun, WANG Xiao-qi, CHENG Shui-yuan* (Key Laboratory of
Beijing on Regional Air Pollution Control, Faculty of Environmental and Life, Beijing University of Technology, Beijing 100020,
China ) China Environmental Science
Abstract: Quantitative analysis of meteorological and human factors during the epidemic are important means to effectively
evaluate the causes of air quality changes in different research areas. This study selected Xingtai City, Hebei Province as the
research object, took 2020 epidemic situation as an experimental scenario of extreme emission reduction under the extreme control
measures, and 2021 epidemic situation as an experimental analysis scenario of future normalized epidemic prevention and control.
Compared with the previous period of the epidemic, the