基于改进时间序列统计模型的空气质量预报  

Modeling of Optimization Time Series Forecasting and Its Application to Prediction of Atmospheric Quality

作　　者：孙暠[1] 宁平[1] 史建武[1] 张朝能[1] 钟曜谦 孙宝磊[1] SUN Hao NING Ping SHI Jianwu ZHANG Chaoneng ZHONG Yaoqian SUN Baolei(Faculty of Environmental Science and Engineering, Kunming University of Science and Technology, Kunming 650093, Chin)

机构地区：[1]昆明理工大学环境科学与工程学院,云南昆明650093

出　　处：《昆明理工大学学报（自然科学版）》2017年第1期91-97,共7页Journal of Kunming University of Science and Technology(Natural Science)

基　　金：国家自然科学基金项目(21207055);云南省社会发展科技计划项目(2012CA016)

摘　　要：针对复杂时序数据的预报,提出了自回归的一般近似方式以及动态最小二乘的预报方法.利用带约束的最优化思想,在这种动态时间序列预报方式的基础上建立了能够综合统计时间序列模型与机理决定模型预报模型,和动态的参数估计方式.模型对采自玉溪市大营街大气质量监测数据进行了预报,预报结果与实测吻合较好.模型的参数估计方式简单,能够综合机理模型和统计模型.其在对含有不确定性因素影响的复杂问题的预报具有推广意义.Time sequence data,such as dynamic atmospheric quality of monitoring data,is affected by various uncertainties and decisive rules in physics. The complex air pollutant density depends on the mechanisms of mass transfer and random impacts such as turbulence. Models applied to forecasting complex time series data should reflect the two kinds of impacts mentioned above. In this paper,two forecasting models are built based on statistical time series theory,which has been successfully applied to uncertainty analysis. The method of optimization programming is employed in the modeling of the forecasting plan. It is argued that,any other necessary decisive mechanic models could be added in this plan,while random noises are taken into account at the same time.Forecasting with this model on the atmospheric quality monitoring time series was carried on. Comparative data set is collected in Dayingjie Yuxi city,Yunnan Province,China. Results generated from this model well fit the data. Thus,it is proposed that models adopted by this paper with the traits of concise parameter estimation algorithm and random-decisive integrated prediction solution can be applied to different complex problems in terms of complex time series forecasting.

关 键 词：时间序列 动态参数估计 综合预报 大气质量监测 

分 类 号：X511[环境科学与工程—环境工程]