基于BP神经网络高风险隧道围岩判定模型的研究

针对铁路高风险隧道地质围岩复杂性、突变性的问题,基于BP神经网络构造了以岩体完整性系数、岩体质量指标、结构面强度系数、岩石单轴饱和抗压强度和地下水渗流量5个参数,进行高风险隧道围岩级别判定模型的研究。通过该模型对相关数据进行学习、训练以及回判和预测,实现判定围岩级别和实际围岩级别基本一致。

Study on Discriminant Model of Rock Surrounding High Risk Tunnel Based on BP Neural Network

In view of the rock surrounding high risk railway tunnel being of complexity and mutagenicity, the 5 parameters, including rock integrity coefficient, rock quality index, strength coefficient of structural plane, saturated uniaxial compressive strength of rock and groundwater seepage discharge, were constituted on the basis of BP neural network to study the model of surrounding rock classification for high risk tunnel. With this model, the related data were studied and trained for discrimination and prediction, which allows the result from the discrimination to agree with the actual classification basically.