| 气象因子对太原地区空气质量的影响研究 |
| |
| 引用本文： | 卢盛栋.气象因子对太原地区空气质量的影响研究[J].新疆气象,2021,15(2):98-105. |
| |
| 作者姓名： | 卢盛栋 |
| |
| 作者单位： | 山西省气象灾害防御技术中心;山西省气象科学研究所 |
| |
| 基金项目： | 国家自然科学基金项目(41805111);山西省气象局科学技术面上项目(SXKMSFW20205222)。 |
| |
| 摘 要： | 利用2017—2019年太原小店区气象与空气质量指数逐时数据,分析了气象因子对空气质量指数的影响,对易污染月份空气质量进行了深入分析,利用神经网络方法构建了气象因子与空气质量指数的关系模型,并与逐步回归模型进行了对比。结果表明:(1)2017—2019年太原地区空气质量6个等级均出现过,其中,良占比最大,为51.9%;严重污染等级最少,为1.9%。每年8月空气质量最好,1月空气质量最差,11月—次年2月为易污染月份。(2)气温、风速、气压对空气质量指数的影响主要表现为负相关,9月—次年2月呈显著负相关;降水主要表现为负相关;相对湿度对空气质量指数影响较大,8月—次年3月呈显著正相关,6—7月呈显著负相关。(3)易污染月份,太原地区空气质量以良为主,占40.7%;其次为轻度污染,占24.8%;严重污染占4.6%。相对湿度、风速对空气质量影响较大,表现为强相关。(4)神经网络构建的气象因子与空气质量指数的关系模型,与逐步回归分析模型相比,准确率由91%提高到94%。神经网络模型模拟效果更佳,为太原地区治理大气污染提供重要的参考价值。
|
| 关 键 词： | 空气质量 气象因子 逐步回归 神经网络 |
The Influence of Meteorological Factors on Air Quality in Taiyuan Area |
| |
| LU Shengdong,LI Fen,CAI Zhaoxin,LI Qiang,NIU Yongbo,CHEN Ling,HE Jieying.The Influence of Meteorological Factors on Air Quality in Taiyuan Area[J].Bimonthly of Xinjiang Meteorology,2021,15(2):98-105. |
| |
| Authors: | LU Shengdong LI Fen CAI Zhaoxin LI Qiang NIU Yongbo CHEN Ling HE Jieying |
| |
| Institution: | (Shanxi Meteorological Disaster Prevention Technology Centre,Taiyuan 030012,China;Shanxi Institute of Meteorological Science,Taiyuan 030002,China) |
| |
| Abstract: | Based on the hourly meteorology and air quality index data of Xiaodian district in Taiyuan from 2017 to 2019,the influence of meteorological factor on air quality index was analyzed.Further study of the air quality in the months prone to pollution was carried out.The relationship model between meteorological factors and air quality index was built by using the method of neural network,and compared with stepwise regression model.The results show that:(1)From 2017 to 2019,all the six levels of air quality grades were appeared in Taiyuan,with the highest ratio(51.9%)of good grade and the lowest ratio(1.9%)of serious pollution grade.The air quality was the best in August and the worst in January.The pollution-prone months were from November to February of the next year.(2)Air temperature,wind speed and air pressure had a negative correlation with the air quality index,especially from September to February of the next year.Precipitation was mainly negatively correlated to air quality index too.Relative humidity had a great influence on the air quality index,which was significantly positively correlated from August to March of the following year,and significantly negatively correlated from June to July.(3)In the pollution-prone months,the air quality in Taiyuan was mainly good with accounting for 40.7%,followed by light pollution with accounting for 24.8%,and the severe pollution accounted for 4.6%.Relative humidity and wind speed had great influence on air quality,showing a strong correlation.(4)Compared with the stepwise regression analysis model,the accuracy of the neural network model on the relationship between meteorological factors and air quality index increased from 91%to 94%.The simulation effect of neural network model is better,which provides an important reference for air pollution control in Taiyuan. |
| |
| Keywords: | air quality meteorological factor stepwise regression neural network |
| 本文献已被 CNKI 维普 等数据库收录！ |
| 点击此处可从《新疆气象》浏览原始摘要信息 |
| 点击此处可从《新疆气象》下载免费的PDF全文 |
|
