基于ARIMA模型的蚌埠市PM2.5预测

本文首先从真气网选取蚌埠市2018年6月1日至2019年5月31日的PM2.5日浓度数据作为样本数据,接着运用EVIEWS10.0软件并借助由Box和Jenkins创立的ARIMA(p,d,q)模型对样本数据进行合理建模,并验证所建立的AR(1)模型的合理性;然后运用所建立的AR(1)模型对蚌埠市2019年6月2日至2019年6月6日的PM2.5日浓度进行预测,最后将预测结果与实际值进行比较,结果表明:使用建立的AR(1)模型对蚌埠市短期内PM2.5浓度预测所得值与蚌埠市PM2.5浓度的实际值相对误差较小,其误差大小在10%之内,适合做蚌埠市PM2.5日浓度的短期预测。

PM2.5 Forecast of Bengbu City Based on ARIMA Model

This paper selects PM2.5 concentration data in Bengtbu City from June 1,2018 to May 31,2019 as sample data for the analysis of time series,uses EVIEWS 10.0 software and ARIMA model created by Box and Jenkins to model the sample data and checks its rationality,and then uses AR(1)model from June 2,2019 to May 31,2019 to verify its rationality;the concentration of PM2.5 was foreseen in June 6,19 and the expected results were compared to the actual values.The results showed that the relative error between the expected values and the actual values was small,the error is less than 10%,which was appropriate for the short-term prediction of the PM2.5 concentration in Benbu City.