Landslide displacement non-parametric probability density prediction method and device and storage medium

Abstract

The invention provides a landslide displacement non-parametric probability density prediction method and device, and a storage medium. The prediction method comprises the following steps: obtaining landslide deformation induction factors and landslide displacement historical monitoring data within a preset time; utilizing a Gaussian kernel density estimation method to obtain an edge distribution function of landslide deformation inducing factors, adopting a correlation structure analysis function to evaluate the correlation degree between each landslide deformation inducing factor and landslide displacement, and optimizing the landslide deformation inducing factors to determine input variables of a prediction model; building a prediction model between the final input variable and the landslide displacement through a kernel function quantile regression support vector machine, and obtaining a landslide displacement prediction result corresponding to the prediction model; and performing non-parametric kernel density estimation on each landslide displacement prediction result by using a kernel density estimation method. The technical scheme provided by the invention has the beneficialeffects that the uncertainty in landslide prediction can be measured, complete prediction value probability density distribution is given, and the reliability of landslide displacement prediction canbe improved.

Description

technical field [0001] The invention relates to the technical field of landslide displacement prediction, in particular to a non-parametric probability density prediction method, equipment and storage medium for landslide displacement. Background technique [0002] Landslides are one of the main types of geological disasters in mountainous areas, which seriously threaten people's lives and property safety. Statistics show that landslides worldwide cause billions of dollars in economic losses and thousands of casualties every year. The non-parametric probability density prediction of landslide displacement can predict the future development trend of landslide deformation, so as to actively take disaster prevention measures and countermeasures, which is one of the most efficient and economical measures to reduce landslide disaster losses. [0003] Since the Japanese scholar Saito proposed the creep three-stage model, a large number of landslide prediction models have sprung u...

AI Technical Summary

Problems solved by technology

[0004] In view of this, the embodiment of the present invention provides a non-parametric probability density prediction method, equipment and storage medium for landslide displacement, aiming to solve the problem that existing prediction methods ignore the randomness and fuzziness characteristics of the landslide system, and ignore the input There are many uncertain factors such as the uncertainty of the model, the uncertainty of the model, and the uncertainty of the parameters, which have certain limitations in theory.

Images