Ground temperature quality control method based on genetic algorithm and moving surface fitting

Abstract

The invention discloses a ground temperature quality control method based on a genetic algorithm and moving surface fitting. The ground air temperature quality control method GA-DCR based on the genetic algorithm and moving surface fitting meets the quality control requirements of ground air temperature observation data in China, and suspicious data can be effectively discriminated; the GA-DCR method shows error detection capability, adaptability and stability superior to those of an IDW method and an SRT method in different regions nationwide; the main reason is that the GA-DCR method adoptsa cosine value and a root-mean-square error value to analyze the correlation between the temperature observation data of the adjacent stations and the temperature observation data of the target station, and the parameters are optimized through the genetic algorithm, and an optimal experiment reference site is obtained; the site with the high correlation between the data is selected as the reference site, and the COS and the RMSE are utilized to assign the weight to the reference site, so that the influence of landform on the quality control algorithm is reduced, and the accuracy of the GA-DCRmethod is improved.

Description

technical field [0001] The invention relates to the technical field of surface air temperature quality control, in particular to a surface air temperature quality control method based on genetic algorithm and moving surface fitting. Background technique [0002] In the context of global informatization, numerical weather prediction has become a necessary means of weather forecasting that affects economic development and social progress. Scholars at home and abroad have conducted in-depth research on numerical weather prediction through a large number of observation experiments and numerical simulations. The research shows that data assimilation technology is a prerequisite for ensuring the accuracy of numerical weather prediction, and quality control of meteorological data is a necessary part of assimilation. In recent years, scholars at home and abroad have conducted extensive research on the quality control algorithms of meteorological observation data, from the observatio...

AI Technical Summary

Problems solved by technology

The selection and weighting of reference stations by these traditional quality control algorithms are basically determined by the Euclidean distance from the target station, and the stations closer to the target station are selected as reference stations and given higher weights. Weighting works better for densely distributed sites and plain areas, but poorer for complex terrain and sparsely sited areas

[0003] The traditional quality control method selects all meteorological stations in the designated area around the target station as reference stations, and the actual temperature value of the meteorological stations closer to the target station may be affected by terrain and other factors. The distance to judge the degree of correlation leads to a larger weight given to the site, which ultimately affects the quality control results

Images