【摘要】 利用人工神经网络技术,建立了围岩压力预测的BP神经网络模型,并以数值模拟的计算结果作为实测围岩压力的控制指标,采用围岩压力的实测数据对网络进行了训练,最后以此训练好的BP神经网络对围岩压力进行了预测。通过与非线性预测对比表明,该人工神经网络模型具有较高的预测精度,为预测围岩压力提供了一种新的方法。更多还原

【Abstract】 The surrounding rock pressure of highway tunnel is an important index to assess its stability and economy of support structure in NATM.Based on data column which is related to measurement time sequence,we can set up some effective models and methods to predict the surrounding rock pressure.According to the surrounding rock pressure characteristics of highway tunnel,this paper introduces BP neural network to establish the model of surrounding rock pressure prediction.Pressure cells were installed and monitoring data of surrounding rock pressure were adopted to train BP neural network.Finally,the BP neural network was applied to predict surrounding rock pressure.By comparing the results with those from non-linear method,it indicates that the model can give high prediction precision,which provides a new way for prediction of surrounding rock deformation.更多还原