基于GM(1,1)-PCA的环境预测与分析研究  

Research on environment prediction and analysis based on GM(1,1)-PCA

作　　者：贺金龙 吴晟[1] 周海河[1] 李英娜[1] 吴兴蛟 李天龙[1] 马颢瑄 

机构地区：[1]昆明理工大学信息工程与自动化学院,昆明650500

出　　处：《信息技术》2018年第1期105-109,共5页Information Technology

摘　　要：为了预测北京市环境污染情况,首先运用灰色系统理论建立污染物GM(1,1)预测模型,其次使用残差检验模型合理性,最后采用主成分分析方法对模型数据进行分析。分析模型预测的结果,可以得知未来5年,北京市环境污染状况呈现逐年好转的趋势,但环境指标化学需氧量和区域环境噪声平均值呈现逐年递增的趋势。主成分分析可得出未来5年,北京市的最主要污染物是有机污染物、道路交通干线噪声和二氧化硫。使用灰色系统预测模型与主成分分析方法结合,不仅能预测未来几年各污染物的具体值,而且还能分析出最主要的污染物。In order to predict the environmental pollution situation in Beijing,first the grey system theory is used to set up pollutants GM（ 1,1）,then the residual is used to test whether the model is rational or not,at last the principal component analysis method is used to analyze the model data. Analyzing the results of prediction model,it is known that： in the next five years,Beijing environment pollution condition presents the tendency of improved year by year,but environmental indicators of chemical oxygen demand（ cod） and the regional environmental noise average shows the tendency of increasing year by year. Principal component analysis can be concluded that in the next five years,the main pollutants in Beijing are organic pollutants,main road traffic noise and sulfur dioxide. Grey system forecasting model with principal component analysis method not only can predict the specific value of each indicator in the coming years,but also can analyze the main pollutants.

关 键 词：灰色系统理论 GM(1 1)预测模型 主成分分析 残差检验 模型合理性 污染物 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]