A Risk Prediction Method for tbm Crash Based on Numerical Samples and Random Forest

Abstract

The invention discloses a TBM jamming risk prediction method based on numerical samples and random forests, which includes: establishing a refined numerical simulation model, simulating the time-dependent deformation of surrounding rock based on a creep damage model, and realizing the simulation of the TBM construction process; Set the values ​​of different jamming factors in the numerical simulation model, and calculate numerical samples containing different working conditions; establish jamming risk discrimination indicators, and calibrate the jamming risk level of samples; establish a random forest model, and conduct model training based on numerical samples , use the trained random forest model to predict the jamming risk level of the actual construction section. The present invention builds a machine jamming numerical sample library based on refined numerical simulation, which overcomes the problems of few and unbalanced monitoring samples existing in the application of machine learning in engineering; uses the trained random forest model to quickly predict the risk level of machine jamming in the actual construction section, thereby Guide the prevention and control of disasters in advance to ensure the safe and efficient construction of TBM.

Description

technical field [0001] The invention relates to the technical field of TBM (Tunnel Boring Machine, tunnel boring machine) excavation construction, in particular to a TBM card machine risk prediction method based on numerical samples and random forests. Background technique [0002] With the rapid development of my country's economy, the intensification of infrastructure construction has been driven. In recent years, key projects and projects in the fields of water conservancy, transportation, and mining have all involved the construction of deep and long tunnels. Compared with the drilling and blasting method, TBM has the advantages of fast construction speed, less disturbance to surrounding rock and less disturbance to the ecology, and is more and more widely used in tunnel engineering. Deep buried tunnel projects often face the problems of high ground stress and strong excavation disturbance. The deep rock mass has obvious aging deformation characteristics after excavatio...

AI Technical Summary

Problems solved by technology

[0006] In order to overcome the problem that the jamming risk in the current TBM tunnel construction is difficult to predict quickly and in real time, the present invention proposes a TBM jamming risk prediction method based on numerical samples and random forests

Images