MFDFA支持下的南京市PM2.5时间尺度变化研究

摘要

PM2.5时间尺度变化的研究对人类的生产生活具有重要的指导意义.通过在Matlab软件中运用多重分形消除趋势波动分析法( Multifractal Detrended Fluctuation Analysis,MFDFA)先对南京市空气污染因子PM2.5近一年的数据进行统计处理,再对得出的q阶Hurst指数以及多重分形谱图进行分析.结果表明：南京市城区PM2.5浓度总体上呈现先下降后上升的趋势,可见南京以小时为尺度的PM2.5浓度具有长程相关性,并且存在多重分形现象；对于山西路和草场门监测站而言,当q取(0,5)时,值均在(0．5,1)范围内,说明了PM2.5浓度均有长程相关性,其未来整体PM2.5浓度的变化趋势将继续上升.%The research on PM2. 5 time scale change is of great significance to production and life. By the MATLAB software, multifractal detrended fluctuation analysis method ( MF-DFA method ) was used to count and process nearly a year’s PM2. 5 data in Nanjing, which was one of air pollution factors, and then analyze the q Hurst exponent and multifractal spectrum. The results showed that the PM2. 5 concentration had a trend of decreasing at first and then rose in the urban area of Nanjing, indicating that there was a long range correlation with the PM2. 5 concentra-tion on the hourly scale in Nanjing, and multi fractal phenomenon existed. It was forecast that there would be an in-creasing trend of the PM2. 5 concentration in the urban area of Nanjing in the next year.