Inverse analysis of mechanical parameters of surrounding rock based on BP neural network optimization
Nie Rongshan
Abstract：In order to solve the problem of poor accuracy and high difficulty in determining rock mass parameters under complex geological structure,taking the roadway of 2-1051 working face as the re-search object,the inverse analysis technology of mechanical parameters of surrounding rock based on BP neural network optimization was proposed.The measured displacement data of roadway monitoring points were used as the comparison value,and the elastic modulus,internal friction angle and cohesion were determined as the inversion parameters through parameter sensitivity analysis,and the BP neural network structure was constructed,and the parameter samples were trained and opti-mized by matlab toolbox.The elastic modulus is determined to be 0.824 6 GPa,the internal friction angle is 16.231 7°,and the cohesion is 0.735 9 MPa.The test results are basically consistent with the in-situ measurement data of the rock mass on site.When the inversion results are introduced into the numerical model and the displacement change data obtained through forward analysis are compared with the measured data,they are also basically consistent.The inversion accuracy is relatively high and can meet the engineering prediction requirements,thus having certain application and promotion value.
Keywords：BP neural networknumerical simulationdisplacement inverse analysismechanical parame-terssensitivity
Publication Date：2026-02-28
Online Publishing Date：2026-01-31(First online date of this platform, not the publication date of the document)
Pages：5( 43-47 )