基于空间插值方法的路面结构设计湿度指标精度研究

Study on the Accuracy of Pavement Structure Design Based on Spatial Interpolation Methods

目的选取最优的空间插值方法求取年降水量,为沥青路面结构组合设计提供湿度参数,进而达到依据区域湿度特点规范设计道路结构参数的目的.方法采用4种半变异函数模型的普通克里格法等3种常见的空间插值方法对辽宁省年降水量空间要素进行插值研究,采用交叉验证法评价各方法的插值效果.结果就辽宁省年降水量而言,4种半变异函数模型中的环形模型精度最高.利用CM半变异模型的普通克里格法和其他两种插值方法计算所得平均绝对误差、平均相对误差和均方根误差具有一致性,其中OK法精度最高.结论选取CM半变异模型的OK法求取年降水量可满足计算要求,进而为道路结构设计的湿度参数选取提供参考.

The most precise spatial interpolation method is used to calculate the annual precipitation and provides the humidity parameters to the designers when the structures of asphalt pavements and then gets the target of designing the structures of asphalt pavements depended on humidity charac- teristics regionally and standard. In this paper ,the ordinary kriging（OK） which applies four semi- variogram models and other two common spatial interpolation methods of precipitation to interpo- late precipitation spatially are used respectively. Meanwhile, by the method of crossing validation evaluation interpolation effects of three spatial interpolation methods, the results are that CM have the highest accuracy among four semivariogram models of annual precipitation in Liaoning prov- ince. At the same time ,by OK method with CM semivariogram and other two spatial interpolation methods interpolating precipitation spatially indicates that there are consistence in mean absoluteerror,mean relative error and root mean square error of three spatial interpolation methods. Be- sides, OK has the highest accuracy among three spatial interpolation methods. In conclusion, calcu- lating the precipitation which is out of meteorological stations or in the distribution of scarce lands in OK method with CM semivariograrn can meet the purpose of research in Liaonin- orovince.