时间序列分析在大气环境中的应用  

APPLICATION OF THE TIME SERIES ANALYSIS IN THE ATMOSPHERIC ENVIRONMENT

作　　者：樊敏[1] 顾兆林[1] 

机构地区：[1]西安交通大学人居环境与建筑工程学院,西安710049

出　　处：《环境工程》2009年第S1期540-542,共3页Environmental Engineering

摘　　要：以西安市2004—2008年PM10月平均浓度时间序列为例,应用小波分解和重构对PM10浓度时间序列的变化进行了分析,得到了PM10的年变化趋势和突变特征。将时间序列平稳化预处理后,根据时间序列的自相关和偏自相关图进行判断,得到合适的时间序列模型ARIMA(1,1,1)(2,0,0)S,然后进行参数估计得到模型的各个参数。利用得到的模型对2009—2010年西安市PM10浓度值进行预测,研究结果表明,将小波分析和ARIMA(1,1,1)(2,0,0)S应用于大气污染物浓度时间序列的分析是可行的。In illustration of the PM_ 10 time series of monthly average concentration from 2004 to 2008 in Xi'an City,the yearly change trend of PM_ 10 time series and jump features of the variations are analyzed using the wavelet decomposition and reconstruction.The paper makes the data steady,and utilizes the character of the autocorrelation and partial autocorrelation of the time series in order to obtain the suitable time series model ARIMA(1,1,1)(2,0,0)S.Through the model it is predicted the values of PM_ 10 in Xi'an from 2009 to 2010.The results show that the wavelet analysis and the model ARIMA(1,1,1)(2,0,0)S can clearly demonstrate the features of air pollutant time series of concentration.

关 键 词：PM10 时间序列 小波分析 ARIMA(1 1 1)(2 0 0)S模型 预测 

分 类 号：X831[环境科学与工程—环境工程]