Predicting Roadway Surrounding Rock Deformation and Plastic Zone Depth Using BP Neural Network

Abstract

It is a complex nonlinear problem to analyze the interaction between bind and the deformation of surrounding rocks and the depth of plastic zones as well as the deformation caused by destruction. This study tackles this problem by taking into account the size of roadway section, burial depth, rock parameters and the in homogeneity of surrounding rock. The software, namely--FLAC has been used perform the numerical analysis. The sample data that affect the deformation and plastic zone size of the surrounding rock have been investigated. The BP Neural Network Predictive Model has been developed to solve the nonlinear mapping issue. With the strength of the artificial neural network, the prime deformation and the plastic zone range of the surrounding rocks caused by the roadway excavation were obtained. The developed predictive model has further been tested with a case study on the roadway with the Jalainur northern iron ore. The predictions of the ore’s deformation and plastic zone range of surrounding rock are found to be consistent with the field observations.