基于空间插值的室内湿度场模拟方法比较分析
|
摘要
空间插值是研究空间因子区域分布特征的重要方法,可实现空间因子可视化展示。不同的空间因子受影响不同,导致空间插值方法的可靠性和适用性存在差异。为了确定室内区域湿度场动态实时化模拟的最佳方法,以Inter实验室内54个温湿度传感器的一个月采集数据为研究案例。综合分析了反距离权重插值、径向基函数插值、普通克里金插值、协同克里金插值、时空克里金插值和时空协同克里金插值6种空间插值方法,并采用交叉验证的方法对插值结果进行比较。结果表明,时空协同克里金插值方法在模拟准确度、可信度和反应极值等能力都要优于其余几种插值方法。
Spatial interpolation is an important method to study the spatial distribution characteristics of spatial factors,which can realize the visual display of spatial factors. Different spatial factors are affected differently,which leads to differences in the reliability and applicability of spatial interpolation methods. In order to determine the best method for dynamic real-time simulation of indoor humidity field,this paper takes the monthly data collected from 54 temperature and humidity sensors in the Inter laboratory as the study case. Six spatial interpolation methods,including inverse distance weight interpolation,radial basis function interpolation,ordinary kriging interpolation,coordinative kriging interpolation,spatial-temporal kriging interpolation and spatial-temporal coordinative Kriging interpolation,are comprehensively analyzed and the cross-validation method is used to compare the interpolation results. The results show that the spatial-temporal collaborative kriging interpolation method is superior to the other interpolation methods in terms of simulation accuracy,credibility and reaction extremum.
Spatial interpolation is an important method to study the spatial distribution characteristics of spatial factors,which can realize the visual display of spatial factors. Different spatial factors are affected differently,which leads to differences in the reliability and applicability of spatial interpolation methods. In order to determine the best method for dynamic real-time simulation of indoor humidity field,this paper takes the monthly data collected from 54 temperature and humidity sensors in the Inter laboratory as the study case. Six spatial interpolation methods,including inverse distance weight interpolation,radial basis function interpolation,ordinary kriging interpolation,coordinative kriging interpolation,spatial-temporal kriging interpolation and spatial-temporal coordinative Kriging interpolation,are comprehensively analyzed and the cross-validation method is used to compare the interpolation results. The results show that the spatial-temporal collaborative kriging interpolation method is superior to the other interpolation methods in terms of simulation accuracy,credibility and reaction extremum.
引文
[1] Huang Y,Shen X,Li J,et al. A method to optimize sampling loca-tions for measuring indoor air distributions[J]. Atmospheric Environ-ment,2015,102(102):355-365.
[2] Igarashi K,Dai H,Oguchi K. Estimation for spatial distribution oftemperature and humidity by using sensor networks[C]∥Internation-al Conference on Telecommunications and Signal Processing,2011.
[3]周伟,汪小旵,李永博.可变边界条件下的Venlo温室温度场三维非稳态模拟[J].农业机械学报,2014,(11):304-310.
[4]于洋,卫伟,陈利顶,等.黄土高原年均降水量空间插值及其方法比较[J].应用生态学报,2015,(4):999-1 006.
[5]董志南,郑拴宁,赵会兵,等.基于空间插值的风场模拟方法比较分析[J].地球信息科学学报,2015,(1):37-44.
[6]丁卉,余志,徐伟嘉,等. 3种区域空气质量空间插值方法对比研究[J].安全与环境学报,2016,(3):309-315.
[7]杨顺华,张海涛,郭龙,等.基于回归和地理加权回归Kriging的土壤有机质空间插值[J].应用生态学报,2015,(6):1 649-1 656.
[8]郑鑫,杨涛,师鹏飞,等.缺资料地区日降雨空间插值方法研究[J].中国农村水利水电,2017,(3):13-16.
[9]段平,盛业华,张思阳,等.顾及异向性的局部径向基函数三维空间插值[J].武汉大学学报(信息科学版),2015,(5):632-637.
[10]孙军波,符国槐,李清斌,等.南方大棚内气温空间分布与网格化插值的研究[J].浙江农业科学,2015,(12):2 047-2 051.
[11]万博雨,付聪,郑世健,等.智能温室精细化控制中传感器布点的研究[J].节水灌溉,2017,(10):74-78.
[12] Li J,Heap A D. Spatial interpolation methods applied in the envi-ronmental sciences[M]. Elsevier Science Publishers B V,2014.
[13]胡丹桂,舒红.基于协同克里金空气湿度空间插值研究[J].湖北农业科学,2014,(9):2 045-2 049.
[14]许美玲,邢通,韩敏.基于时空Kriging方法的时空数据插值研究[J].自动化学报,2018,44:1-8.
[15]胡丹桂,舒红,胡泓达.时空CoKriging的变异函数建模[J].华中师范大学学报(自然科学版),2015,(4):596-602.
[16]修思玉,吕泓辰,康世伦,等.地理信息系统中不同空间插值方法的比较研究[J].森林工程,2014,(5):110-113.
[2] Igarashi K,Dai H,Oguchi K. Estimation for spatial distribution oftemperature and humidity by using sensor networks[C]∥Internation-al Conference on Telecommunications and Signal Processing,2011.
[3]周伟,汪小旵,李永博.可变边界条件下的Venlo温室温度场三维非稳态模拟[J].农业机械学报,2014,(11):304-310.
[4]于洋,卫伟,陈利顶,等.黄土高原年均降水量空间插值及其方法比较[J].应用生态学报,2015,(4):999-1 006.
[5]董志南,郑拴宁,赵会兵,等.基于空间插值的风场模拟方法比较分析[J].地球信息科学学报,2015,(1):37-44.
[6]丁卉,余志,徐伟嘉,等. 3种区域空气质量空间插值方法对比研究[J].安全与环境学报,2016,(3):309-315.
[7]杨顺华,张海涛,郭龙,等.基于回归和地理加权回归Kriging的土壤有机质空间插值[J].应用生态学报,2015,(6):1 649-1 656.
[8]郑鑫,杨涛,师鹏飞,等.缺资料地区日降雨空间插值方法研究[J].中国农村水利水电,2017,(3):13-16.
[9]段平,盛业华,张思阳,等.顾及异向性的局部径向基函数三维空间插值[J].武汉大学学报(信息科学版),2015,(5):632-637.
[10]孙军波,符国槐,李清斌,等.南方大棚内气温空间分布与网格化插值的研究[J].浙江农业科学,2015,(12):2 047-2 051.
[11]万博雨,付聪,郑世健,等.智能温室精细化控制中传感器布点的研究[J].节水灌溉,2017,(10):74-78.
[12] Li J,Heap A D. Spatial interpolation methods applied in the envi-ronmental sciences[M]. Elsevier Science Publishers B V,2014.
[13]胡丹桂,舒红.基于协同克里金空气湿度空间插值研究[J].湖北农业科学,2014,(9):2 045-2 049.
[14]许美玲,邢通,韩敏.基于时空Kriging方法的时空数据插值研究[J].自动化学报,2018,44:1-8.
[15]胡丹桂,舒红,胡泓达.时空CoKriging的变异函数建模[J].华中师范大学学报(自然科学版),2015,(4):596-602.
[16]修思玉,吕泓辰,康世伦,等.地理信息系统中不同空间插值方法的比较研究[J].森林工程,2014,(5):110-113.