基于遗传算法和克里金的气温站网拓展设计

Expansion Design of Temperature Station Network Based on Genetic Algorithm and Kriging

利用江苏省2016年58个地面气象站日均温数据,针对其站点布局的不合理以及对未来站网的拓展,提出一种基于地统计克里金法和遗传算法的方法,以研究区域现有地面气温站点,利用克里金方法计算整个区域的克里金方差并对研究区域进行估值作为模拟站点,通过遗传算法从模拟站点中寻找最优的站网组合,解决局部最优站点组合起来不一定是整体最优的问题。最终确定了6个添加站点位置,研究结果为未来气温站网的建设提供了理论参考。

The daily average temperature data of 58 ground-based meteorological stations in Jiangsu Province in 2016 was taken as samples.To deal with the unreasonable layout of the site and the expansion of the future station network,a method based on geostatistical Kriging method and genetic algorithm was proposed.The method was used to study the existing ground-based meteorological stations.The Kriging method was used to calculate the Kriging standard error for the entire region and evaluate these regions as simulation sites.The optimal network combination from the simulation site was calculated through genetic algorithm to solve the local optimal site.It solved the problem of the combination of local optimal sites was not necessarily the overall optimal scheme.Finally,six additional site locations were identified,which could provide a reference for the construction of the future temperature station network.