Pile-soil interaction prediction analysis method based on machine learning

Abstract

The invention discloses a pile-soil interaction prediction analysis method based on machine learning, and belongs to the technical field of foundation engineering. The method comprises the following steps: establishing a pile-soil variable parameter sample by adopting a Latin hypercube sampling method, modeling the parameter sample by adopting a numerical simulation method to obtain a stress deformation value of a pile body corresponding to the parameter sample, and performing sensitivity analysis on an input variable and a demand variable by adopting a Lasso method to reduce the dimension ofthe input variable; dividing the parameter samples into K parts with equal quantity for cross validation; establishing a BP neural network model based on an L-M algorithm, defining the number of neurons of the hidden layer within a certain range for cyclic traversal operation, determining the optimal number of neurons of the hidden layer by comparing training errors, and predicting the stress deformation of the pile body by using the trained neural network model. The method has the advantages of being clear in analysis process, high in reliability and high in efficiency, and a theoretical basis is provided for design and application of the pile foundation.

Description

technical field [0001] The invention relates to a method for predicting and analyzing pile-soil interaction based on machine learning, which belongs to the technical field of foundation engineering. Background technique [0002] Pile foundations are widely used in various engineering constructions, and are the most commonly used treatment methods for bad foundations. With the development of pile foundation technology, some new pile foundation technologies have appeared, such as spiral steel pile, composite geotechnical material encapsulated bulk pile, etc. There are many factors influencing the design of these piles. Understanding its stress and deformation characteristics is of great significance to the safety, stability and economy of the project. The bearing capacity of the pile foundation is the result of the joint action of the pile and the soil. The traditional field load test is the most commonly used analysis method. By analyzing the bearing capacity of a single pil...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art, to provide a fast and accurate pile-soil interaction prediction analysis method based on machine learning, to find out the stress and deformation characteristics of piles with complex influencing factors, and to solve the problem of traditional analysis. The problem of low efficiency and high cost of the method provides a theoretical basis for the design and application of piles

Images