Landslide interval prediction method based on machine learning and probability theory

Abstract

The invention relates to a landslide interval prediction method based on machine learning and a probability theory, and the method comprises the following steps: decomposing landslide accumulative monitoring displacement time series data into a plurality of IMF subitems and a residual item R through employing a VMD decomposition technology, superposing the IMF subitems to obtain a displacement fluctuation item, and taking the R as a displacement trend item; polynomial fitting is carried out on the displacement trend term, and a least square method is adopted to predict the displacement trend term; selecting an effective influence factor of a displacement fluctuation item according to the Copula model, taking the obtained effective influence factor as an input parameter, taking the obtained displacement fluctuation item as an output parameter, establishing a KELM model, and predicting the displacement fluctuation item; respectively calculating landslide displacement prediction errors of the displacement trend item and the displacement fluctuation item; determining a prior distribution probability density function which most accords with the real distribution of each displacement item, solving a corresponding cumulative probability distribution function according to the determined prior distribution probability density function, and calculating to obtain a parameter method prediction interval corresponding to the predicted moment of each displacement item.

Description

technical field [0001] The invention relates to a landslide data processing method, in particular to a landslide interval prediction method based on machine learning and probability theory. Background technique [0002] Landslides are a common natural geological disaster, especially in the mountainous areas of southwest my country, which pose a huge threat to the safety of people's lives and properties in the region. Experts and scholars at home and abroad have carried out various researches on the prevention and control of landslides. Practice has proved that the prediction and prediction of landslide displacement changes is still one of the most effective means of landslide disaster early warning. In recent years, the research on landslide displacement prediction has developed from the initial physical model to the later abstract model, and after continuous in-depth research, the abstract model has developed from a deterministic model to a displacement-time series statisti...

AI Technical Summary

Problems solved by technology

[0003] The present invention aims at improving the accuracy of the prediction results and neglecting the research on the reliability of the prediction results in the current landslide prediction research, and introduces a landslide displacement interval prediction method based on the combination of machine learning and probability statistics theory into the landslide prediction science, The aim is to scientifically and reasonably quantify the reliability of landslide displacement prediction results, so as to provide more valuable reference information for early warning and prevention of landslide disasters

Images