【摘要】 以成都市2020年的日均空气质量数据和地面气象数据为基础，对PM_2.5浓度的年际变化和季节分布进行分析，得出成都市2020年较2019年相比PM_2.5平均浓度下降5.94μg/a·m³，年度最低值为4μg/m³;PM_2.5随季节变化规律为：冬高夏低，冬季波动大，夏季稳定，平均浓度冬季最高，秋季次之，夏季最低，并且冬季有26天超出《环境空气质量标准》二级标准；气象因子中，最大持续风速、最高温度、降雨量、能见度等各因子都与PM_2.5呈现出负相关，然后通过建模适用性分析后，对数据样本建立多元线性回归和ARIMA模型，ARIMA模型效果较差，多元线性回归模型准确度较高，适用性强，效果更好，具有较高的应用价值。更多还原

【Abstract】 Based on the daily average air quality data and ground meteorological data of Chengdu in 2020,the interannual variation and seasonal distribution of PM_2.5 concentration were analyzed.It was concluded that the average PM_2.5 concentration in Chengdu in 2020 decreases by 5.94μg/a·m³ compared with that in 2019,and the annual lowest value was 4μg/m³.The seasonal variation of PM_2.5 was as follows:it was high in winter and low in summer,large fluctuation in winter,stable in summer,the average concentration is highest in winter,followed by autumn,and lowest in summer.In addition,26 days in winter exceed the secondary standard of Ambient Air Quality Standard.Among the meteorological factors,the maximum sustained wind speed,maximum temperature,rainfall,visibility and other factors were negatively correlated with PM_2.5.Through the establishment of multiple linear regression model and ARIMA model for data samples,and their applicability analysis,it could be seen that ARIMA model had poor effect,while multiple linear regression model had higher accuracy,strong applicability,better effect and higher application value.更多还原