Efficient rock and earth mass parameter random field modeling method

Abstract

The invention discloses an efficient rock and earth mass parameter random field modeling method. The method comprises the following steps of 1, determining a probability model and statistical characteristics for representing spatial variability of rock-soil body parameters to be simulated; 2, determining the dimension and the scale of a to-be-simulated site; 3, determining the irrelevant distanceof the parameters; 4, gridding and discretizing the to-be-simulated site; 5, determining a covariance matrix based on an irrelevant distance concept; 6, determining a Gaussian random field model; andstep 7, carrying out non-Gaussian random field simulation based on equal probability conversion. Random field modeling is performed based on the improved covariance matrix factorization method, the advantages of the covariance matrix factorization method in random field modeling precision can be brought into play, and random field simulation of any scale can be efficiently achieved on the premisethat a large amount of calculation investment is not needed.

Description

technical field [0001] The invention relates to the technical field of geotechnical engineering, in particular to an efficient random field modeling method of rock and soil mass parameters. Background technique [0002] As a product of nature, rock and soil mass is self-evident in the complexity and diversity of its formation process. Nature has created different rock and soil types in different historical periods, and the differences in the mineral composition and stress history of the same rock and soil lead to different degrees of differences in physical and mechanical parameters, which are called spatial variability. This inherent spatial variability is an important source of uncertainty in geotechnical engineering. Therefore, how to accurately describe the geological model is extremely important for economically and rationally solving the problem of uncertainty in geotechnical engineering. At present, in most geotechnical engineering design methods, soil is regarded as...

AI Technical Summary

Problems solved by technology

The Local Average Subdivision Method (LAS method) has higher modeling efficiency, but the implementation process is more complicated and the simulation accuracy is not high

The advantages of the covariance matrix decomposition method and the Karhunen-Loeve (KL) series expansion method are mainly reflected in the random field modeling process and simulation accuracy, but this method usually requires a large amount of storage space during the modeling process. It is difficult for computers to meet the requirements of 3D random field modeling

Images