KPCA-FOA-LSSVM-based landslide hazard prediction method

Abstract

The invention discloses a KPCA-FOA-LSSVM-based landslide hazard prediction method, which comprises the steps of: firstly, establishing a landslide mass real-time monitoring and early-warning system, acquiring real-time data of a monitoring region, performing standardized processing on the real-time data, and screening main influencing factors of the occurrence of a landslide as input variables byadopting a kernel principal component analysis method; constructing an LSSVM-based landslide hazard forecasting model; secondly, adopting a fruit fly algorithm for parameter optimization, and updatingnetwork parameters; and finally, reconstructing the optimized landslide hazard forecasting model, outputting occurrence grades corresponding to landslide occurrence probabilities, and completing theforecasting. The KPCA-FOA-LSSVM-based landslide hazard prediction method acquires the monitoring data through establishing the landslide mass real-time monitoring and early-warning system, screens themain influencing factors by means of kernel principal component analysis, utilizes a least square vector machine model optimized based on the fruit fly algorithm to train and output the landslide occurrence probabilities, improves the forecast efficiency and increases the precision.

Description

technical field [0001] The invention belongs to the technical field of geological disaster monitoring, and relates to a landslide disaster prediction method based on KPCA-FOA-LSSVM. Background technique [0002] Landslide is a common natural disaster phenomenon, its appearance can destroy farmland and houses, damage roads and water conservancy and hydropower facilities, resulting in power outages, water outages, work stoppages and other consequences, and may even threaten human life in severe cases . Therefore, how to adopt a timely and accurate landslide disaster forecast method has become a hot spot that people pay attention to. [0003] In view of the characteristics of randomness, high frequency and wide range of landslide occurrence, scholars at home and abroad have conducted in-depth research on it, but the methods used have their own advantages and disadvantages. For example, the gray model does not require a large number of training samples and the calculation work...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a kind of landslide disaster prediction method based on KPCA-FOA-LSSVM, solve the problem that the algorithm that existing disaster prediction method adopts is over-learning, inefficiency, and accuracy is not high

Images