Landslide displacement prediction method improved by data assimilation

Abstract

The invention provides a landslide displacement prediction method improved through data assimilation. The landslide displacement prediction method comprises the steps that firstly, calculating a sliding zone integrity index according to an external influence factor-sliding zone integrity index regression model; then dividing the target landslide strip into a plurality of sliding blocks by adoptinga strip division method, calculating the position of a target landslide seepage line according to a Bcissinke equation, and further calculating by adopting a sliding zone shear strength dynamic change index formula to obtain cohesive force and an internal friction angle; then, the internal thrust of a sliding block where the monitoring point is located is calculated through a residual thrust method, and a landslide cumulative displacement value is calculated through a thrust-displacement regression model; and finally, calculating by adopting a particle filter algorithm to obtain an optimal slip band integrity index and an optimal cumulative displacement prediction value. The beneficial effects of the invention are that the method can guarantee that the landslide prediction model does notmake an overshoot response to the severe noise of an external impact factor, improves the prediction precision of the landslide displacement prediction model, and can effectively enable the predictionerror of the landslide displacement prediction model to be converged.

Description

technical field [0001] The invention relates to the field of landslide displacement prediction, in particular to an improved landslide displacement prediction method by using data assimilation. Background technique [0002] At present, the landslide displacement prediction model is mainly realized by intelligent algorithm, dynamic equation and numerical simulation. Among them, the method based on intelligent algorithm mainly uses historical monitoring data to train the model, which is like a "black box" and only cares about its input and output. It is difficult to describe and explain the evolution process and law of landslides through the principles of physical mechanics, and its stability It is not high, and it is easy to make a violent response that does not conform to physical mechanics when the input such as external influence factors contains a lot of noise, resulting in poor applicability. The traditional methods based on dynamic equations and numerical simulation ar...

AI Technical Summary

Problems solved by technology

Among them, the method based on intelligent algorithm mainly uses historical monitoring data to train the model, which is like a "black box" and only cares about its input and output. It is difficult to describe and explain the evolution process and law of landslides through the principles of physical mechanics, and its stability It is not high, and it is easy to make a violent response that does not conform to physical mechanics when the input of external factors and other factors contains a lot of noise, resulting in poor applicability

The traditional methods based on dynamic equations and numerical simulation are mainly limited to the analysis of landslide stability under specific working conditions, because this type of method is a simplified simulation of the evolution process of the landslide, its accuracy is limited and it is difficult to achieve dynamic operation

Images