High slope deformation prediction method and system

Abstract

The invention provides a high slope deformation prediction method and system. The method comprises the following steps: firstly, acquiring historical deformation data of each period of each part of the high slope as sample data; dividing the sample data into a training sample and a test sample; next, using training samples, optimizing a parameter group of a support vector machine model by adoptinga particle swarm algorithm; determining an optimal parameter group of the support vector machine model; obtaining a trained support vector machine model; verifying whether the trained support vectormachine model meets conditions or not by utilizing the test sample; establishing the support vector machine model, re-determining the optimal parameter set of the support vector machine model when thesupport vector machine model does not meet the condition, and finally, predicting the deformation of each part of the high slope by utilizing the support vector machine model meeting the condition, so that the deformation of the high slope is effectively predicted in real time.

Description

technical field [0001] The invention relates to the field of high side slope deformation monitoring, in particular to a high side slope deformation prediction method and system. Background technique [0002] High slope monitoring is a complex monitoring project. There are a lot of fuzzy, complex, uncertain and other external factors. During the work of high slope monitoring, high slope collapse accidents will occur, seriously affecting the construction personnel personal safety and work progress. Therefore, it is of great significance to carry out real-time early warning and monitoring of high slopes and accurately predict the deformation value of high slopes to ensure the personal safety of construction personnel and the progress of the project. [0003] At present, the commonly used high slope deformation prediction methods include statistical regression analysis, BP neural network and so on. Because there are many factors that lead to the collapse of high slopes, the co...

AI Technical Summary

Problems solved by technology

However, the statistical regression model also has its shortcomings and deficiencies, as follows: (1) There are many factors that affect the deformation and seepage of high slopes, and the mathematical model must be optimized in combination with its inherent characteristics and physical causes of monitoring quantities to ensure its regression model. (2) The statistical regression model is only a speculation, which affects the diversity of factors and the unmeasurability of some factors, making regression analysis limited in some cases

The BP neural network model requires a large number of training samples, and the training speed is too slow, which is prone to phenomena such as too slow fitting and convergence.

It can be seen that neither the statistical regression analysis algorithm nor the BP neural network algorithm can effectively predict the deformation of high slopes in real time

Images