Dimension reduction and correction model based on spacial multi-correlation solution set algorithm

Abstract

The invention discloses a dimension reduction and correction model based on a spacial multi-correlation solution set algorithm. The processing method of the model comprises the steps of (1) obtainingdata, wherein the data includes a climate pattern grid data series Xt = (x1, t, x2, t, ..., xm, t) and actual measurement data series Yt= (y1, t, y2, t, ..., yn, t) of each site, m refers to the number of the grids, t refers to time and n refers to the number of the sites; (2) constructing the dimension reduction and correction model based on the spacial multi-correlation solution set algorithm: Y= B + AX + epsilon (epsilon - N (0, sigma 2)), wherein A and B are used for reflecting a related structural relationship between the grid data and the actual measurement data of the sites, and the sigma is a standardized independent random vector; (3) adopting a particle swarm optimization algorithm to figure out the solutions to the coefficient matrix A and the coefficient matrix B in an equation(1). The dimension reduction and correction model based on the spacial multi-correlation solution set algorithm has the advantage of increasing the simulation precision.

Description

technical field [0001] The invention relates to a downscaling correction model based on a spatial multiple correlation solution set algorithm, and belongs to the technical field of statistical downscaling. Background technique [0002] According to the specific information of global climate models, the spatial resolution of the models is low, and most of them are above 1.5°×1.5°. If the model data is directly used in the study of the impact of climate change on hydrology and water resources, it will definitely bring about large errors, and the results will have high uncertainty. In order to study the future climate change in the region, the climate model should be downscaled. [0003] Common downscaling methods include dynamic downscaling, statistical downscaling, and interpolation downscaling. The statistical downscaling method is simple, flexible, easy to implement, and has been widely used in practice. At present, the commonly used statistical downscaling methods (such ...

AI Technical Summary

Problems solved by technology

However (some) predictors and predictor variables may not have a clear physical correlation; and when the correlation between predictors and predictor variables is weak, the error of the downscaling results will be large, and there is a high uncertainty; this downscaling method by introducing multiple predictors may “contaminate” the downscaling results of the predictor variables

In addition, the commonly used statistical downscaling methods can generally only achieve downscaling of a single site, and cannot guarantee the spatial correlation of the downscaling results of each site in the region

Images