基于ANSYS-BP的隧道拱顶沉降预测研究

发布时间：2018-05-03 12:03

本文选题：ANSYS + BP神经网络　；参考：《中国安全科学学报》2014年05期

【摘要】：为控制隧道拱顶下沉过程及其最终下沉量,保障隧道工程建设安全,需要准确预测隧道拱顶沉降量。基于有限元和人工神经网络,结合工程实践,提出ANSYS有限元和BP神经网络相结合(ANSYS-BP)的隧道拱顶沉降预测方法。其预测步骤为:根据隧道地质条件、围岩性质、施工工艺、支护衬砌方式,采用ANSYS建立隧道模型,计算多步工艺施工下隧道拱顶指定单元处沉降值序列,以此构建BP神经网络预测模型,将沉降序列信息作为输入样本对BP神经网络的进行训练、测试。并以某在建公路隧道为研究对象,进行预测分析。结果表明,ANSYS-BP方法能够有效改善隧道拱顶沉降的预测性能,其预测值与实际监测值基本一致,预测精度高于单独使用ANSYS计算或BP网络预测的结果。
[Abstract]:In order to control the process of tunnel dome subsidence and its final subsidence, and to ensure the safety of tunnel construction, it is necessary to accurately predict the tunnel dome settlement. Based on finite element, artificial neural network and engineering practice, a prediction method of tunnel vault settlement is presented based on ANSYS finite element method and BP neural network combined with ANSYS-BP. The prediction steps are as follows: according to the geological conditions of the tunnel, the nature of surrounding rock, the construction technology and the supporting and lining mode, the tunnel model is established by using ANSYS, and the settlement value sequence of the specified unit of the tunnel vault under the multi-step construction is calculated. The prediction model of BP neural network is constructed, and the settlement sequence information is used as input sample to train and test BP neural network. Taking a highway tunnel under construction as the research object, the prediction and analysis are carried out. The results show that ANSYS-BP method can effectively improve the prediction performance of tunnel dome settlement, and its prediction value is basically consistent with the actual monitoring value, and the prediction accuracy is higher than that of ANSYS calculation alone or BP network prediction.
【作者单位】：中南大学资源与安全工程学院;
【分类号】：U452.1

【相似文献】