时空理论变异函数模型及其精度影响
发布时间：2018-12-17 11:37
【摘要】：针对已有的时空模型研究深度不够的问题,该文主要研究两类时空理论变异函数模型,即分离和非分离模型,并分析其对时空预测精度的影响。首先,阐述了两类模型中较为流行的6种时空理论子模型及其参数物理意义;最后以2014年山东省PM2.5日均浓度为例,给出各模型理论表达式,并在Matlab中对各模型运用遗传算法进行参数拟合和时空预测精度验证。实验结果显示,6种模型的整体预测精度顺序与模型拟合精度顺序相当,说明时空理论模型同空间理论模型一样,其模型拟合精度与后续预测精度成正相关。
[Abstract]:In order to solve the problem that the existing spatio-temporal models are not deep enough, this paper mainly studies two kinds of spatio-temporal theoretical variogram models, that is, separated and non-separated models, and analyzes their influence on the accuracy of spatio-temporal prediction. Firstly, six kinds of space-time theory submodels and their parameter physical meaning are introduced, which are popular in the two kinds of models. Finally, taking the daily average concentration of PM2.5 in Shandong Province in 2014 as an example, the theoretical expressions of each model are given, and the parameters fitting and spatio-temporal prediction accuracy of each model are verified by genetic algorithm in Matlab. The experimental results show that the order of the whole prediction accuracy of the six models is the same as the order of the model fitting accuracy, which indicates that the spatio-temporal theoretical model is the same as the spatial theoretical model, and the model fitting accuracy is positively correlated with the subsequent prediction accuracy.
【作者单位】： 华中农业大学资源与环境学院;农业部长江中下游耕地保育重点实验室;
【分类号】：P208
本文编号：2384143
[Abstract]:In order to solve the problem that the existing spatio-temporal models are not deep enough, this paper mainly studies two kinds of spatio-temporal theoretical variogram models, that is, separated and non-separated models, and analyzes their influence on the accuracy of spatio-temporal prediction. Firstly, six kinds of space-time theory submodels and their parameter physical meaning are introduced, which are popular in the two kinds of models. Finally, taking the daily average concentration of PM2.5 in Shandong Province in 2014 as an example, the theoretical expressions of each model are given, and the parameters fitting and spatio-temporal prediction accuracy of each model are verified by genetic algorithm in Matlab. The experimental results show that the order of the whole prediction accuracy of the six models is the same as the order of the model fitting accuracy, which indicates that the spatio-temporal theoretical model is the same as the spatial theoretical model, and the model fitting accuracy is positively correlated with the subsequent prediction accuracy.
【作者单位】： 华中农业大学资源与环境学院;农业部长江中下游耕地保育重点实验室;
【分类号】：P208
【相似文献】
相关会议论文 前1条
1 丁松;党耀国;徐宁;;灰色Verhulst模型的灰导数优化及其应用[A];第25届全国灰色系统会议论文集[C];2014年
相关博士学位论文 前1条
1 刘梦涵;面向特大城市的分层次交通拥堵评价模型及算法[D];北京交通大学;2009年
相关硕士学位论文 前10条
1 曾柯方;几种灰预测模型的参数辨识与优化方法研究[D];西华师范大学;2015年
2 黄巧玲;泾河流域径流预报模型研究[D];西北农林科技大学;2015年
3 刁庆华;面向控制的DOC建模研究[D];吉林大学;2016年
4 倪皓晨;城市道路三维路面模型构建方法研究[D];南京大学;2015年
5 杨知;GM(1,1)模型的优化及其应用研究[D];南京航空航天大学;2009年
6 李辰;交通方式划分的LOGIT模型方法[D];河海大学;2004年
7 钱吴永;数据变换技术与GM(1，1)模型研究[D];南京航空航天大学;2009年
8 朱健;建筑物变形监测数据分析中的灰色改进模型应用研究[D];长安大学;2007年
9 孙禹枫;CAD模型修复技术研究[D];华中科技大学;2007年
10 庞庆非;基于移动变窗口的地表温度空间分辨率提升模型[D];电子科技大学;2013年
，本文编号：2384143
本文链接：https://www.wllwen.com/kejilunwen/dizhicehuilunwen/2384143.html
