| 基于裂区设计的大气能见度拟合及预测模型 |
| |
| 引用本文： | 冯婷,杜锋,刘永红,蔡铭.基于裂区设计的大气能见度拟合及预测模型[J].环境科学与技术,2012(Z1):336-341. |
| |
| 作者姓名： | 冯婷 杜锋 刘永红 蔡铭 |
| |
| 作者单位： | 中山大学智能交通研究中心;Mathematics Department,Tulane University,New Orleans,LA,70118 |
| |
| 基金项目： | 国家自然科学基金项目-基于动态数据驱动的灰霾预报模型的研究(51108471) |
| |
| 摘 要： | 通过收集分析广州能见度及影响因子数据的特征,使用R-project统计软件作为建模工具,采用多层裂区实验构建模型基本框架,选择影响显著的因子进行变量变形处理,对年、月、日、小时的数据特征进行层层分割设计,并在对应层次引入代表时间含义的虚拟变量以提高模型的表征水平,加入二次项、交叉项等提高模型的精度,得到拟合度高、适用性强的能见度拟合及预测模型。
|
| 关 键 词： | 广州 能见度 拟合模型 预测模型 裂区设计 虚拟变量 |
The Visibility Fitting and Forecast Model Based on Split Plot Design |
| |
| FENG Ting,DU Feng,LIU Yong-hong,CAI Ming.The Visibility Fitting and Forecast Model Based on Split Plot Design[J].Environmental Science and Technology,2012(Z1):336-341. |
| |
| Authors: | FENG Ting DU Feng LIU Yong-hong CAI Ming |
| |
| Institution: | 1(1.Intelligent transportation research center of Sun Yat-sen University,Guangzhou 510275,China; 2.Mathematics Department,Tulane University,New Orleans,LA,70118) |
| |
| Abstract: | Through collecting and analyzing the data characteristics and impact factor of Guangzhou visibility,take use of R-project statistical software as the modeling tool,choose multiple split plot design experiment to build the basic framework,make variable treatment to the significant factors,split design the year,month,day,hour of data in layers,introduce dummy variables meaning the time to the corresponding block to raise the level of the model representation,add the quadratic terms,cross terms to improve the accuracy of the model,get the higher fit and applicability model for visibility fitting and forecast. |
| |
| Keywords: | Guangzhou visibility fitting model forecast model split plot design dummy variables |
| 本文献已被 CNKI 等数据库收录！ |
|
