基于SARIMA模型对阿拉尔市大气颗粒物的污染特征分析及预测研究
Analysis and Prediction Research on the Pollution Characteristics of Atmospheric Particulate Matter in Aral City based on SARIMA Model
江宜霖 1王建 1梁朵朵 1张文刚 1杨波 1李煜1
作者信息
- 1. 新疆生产建设兵团生态环境第二监测站,新疆 阿拉尔 843300
- 折叠
摘要
为了解阿拉尔市大气颗粒物PM10、PM2.5的污染状况,便于合理制定大气污染治理措施.本文收集阿拉尔市2017-2023 年PM10、PM2.5的监测数据,构建季节性差分自回归滑动平均(SARIMA)模型,系统性分析了阿拉尔市大气颗粒物PM10和PM2.5浓度的变化特征,并预测未来变化趋势.研究结果表明:阿拉尔市的大气颗粒物PM10、PM2.5 的质量浓度较高,且呈现明显的季节性变化;基于SPSSAU软件自动寻优的大气颗粒物PM10、PM2.5 最优化模型分别为SA-RIMA(0,0,0)(2,0,0)12和SARIMA(1,0,0)(3,0,0)12;运用最优模型对阿拉尔市2023 年大气颗粒物PM10、PM2.5的月平均进行拟合与比对发现,总体相对误差在15%以内,一定程度上反映了模型的拟合效果良好;运用最优模型对阿拉尔市2024 年大气颗粒物PM10、PM2.5质量浓度进行预测发现,预测值与2024 年 1-6 月实测值吻合度较好,进一步表明该模型的准确性.以上研究结果得出,SARIMA(0,0,0)(2,0,0)12和SARIMA(1,0,0)(3,0,0)12 适用于阿拉尔市大气颗粒物的污染状况预测,可为阿拉尔市大气颗粒物污染的环境调控策略提供技术参考.
Abstract
In order to understand the pollution status of atmospheric particulate matter PM10 and PM2.5 in recent years in Alar City,and adjust the air pollution control strategies,this work collected the monitoring data of atmospheric particulate matter PM10 and PM2.5 in Alar City from 2017 to 2023,and constructed a seasonal autoregressive integrated moving average(SARI-MA)model to systematically analyzes the variation characteristics of the concentration of atmospheric particulate matter PM10 and PM2.5 in Alar City,and predict their future trends.The results show that the mass concentrations of atmospheric particu-late matter PM10 and PM2.5 in Alar City are relatively high and exhibit significant seasonal variations.The optimized models for atmospheric particulate matter PM10 and PM2.5,automatically selected by the SPSSAU software,are SARIMA(0,0,0)(2,0,0)12 and SARIMA(1,0,0)(3,0,0)12,respectively.By applying the optimized models to fit and compare the monthly avera-ges of atmospheric particulate matter PM10 and PM2.5 in Alar City in 2023,it was found that the overall relative error is within 15%,to some extent reflecting the good fitting effect of the model.The application of the optimized models to predict the mass concentrations of atmospheric particulate matter PM10 and PM2.5 in Alar City in 2024 revealed a good agreement between the predicted values and the measured values from January to June 2024,further demonstrating the accuracy of the model.The a-bove research results indicate that SARIMA(0,0,0)(2,0,0)12 and SARIMA(1,0,0)(3,0,0)12 can be used to predict the pollution status of atmospheric particulate matter in Alar City,providing technical reference for environmental regulation strate-gies for atmospheric particulate matter pollution in Alar City.
关键词
SARIMA模型/颗粒物/阿拉尔/污染特征/预测Key words
SARIMA model/particulate matter/Alar/pollution characteristics/prediction引用本文复制引用
出版年
2024
NETL
NSTL
万方数据