去噪分析和分离变形预测在深埋隧道中的应用研究

发布时间：2018-01-08 07:20

本文关键词：去噪分析和分离变形预测在深埋隧道中的应用研究　出处：《铁道标准设计》2017年09期 　论文类型：期刊论文

【摘要】：隧道变形易诱发相应的工程问题,对其防治及预测已成为地下工程领域的热点问题。为提高隧道变形的预测精度,达到有效掌握隧道变形规律的目的,以某隧道为工程实例,通过对其监测数据的去噪处理,将隧道变形的原始序列分离为趋势项和误差项序列,并利用GA-BP神经网络和时间序列模型对两序列进行预测,又结合支持向量机模型对前者的预测误差进行修正,以保证预测精度。结果表明:在去噪方面,得出半参数优化卡尔曼滤波的去噪效果最优,其次是sym8小波去噪和奇异谱分析;在预测方面,得出分离预测能一定程度上提高预测精度,但效果不明显,而误差修正模型能很大程度上提高预测精度,综合得到本文预测结果的平均相对误差为1.08%。预测模型具有精度较高等优点,能为深埋隧道的变形预测提供借鉴。
[Abstract]:Tunnel deformation is easy to induce the corresponding engineering problems, prevention and prediction of the tunnel deformation has become a hot issue in the field of underground engineering. In order to improve the accuracy of tunnel deformation prediction, to effectively grasp the tunnel deformation law. Taking a tunnel as an engineering example, the original sequence of tunnel deformation is separated into trend item and error item sequence by de-noising the monitoring data. GA-BP neural network and time series model are used to predict the two sequences, and support vector machine model is used to correct the prediction error of the former to ensure the prediction accuracy. The results show that: in the aspect of denoising. The optimal denoising effect of semi-parametric optimal Kalman filter is obtained, followed by sym8 wavelet denoising and singular spectrum analysis. In the aspect of prediction, it is concluded that the separation prediction can improve the prediction accuracy to some extent, but the effect is not obvious, and the error correction model can improve the prediction accuracy to a great extent. It is concluded that the average relative error of the prediction results is 1.080.The prediction model has the advantages of high accuracy and can be used as a reference for the prediction of deep buried tunnel deformation.
【作者单位】：云南省交通规划设计研究院;四川省冶金地质勘查局水文工程大队;
【分类号】：U456.3
【正文快照】： 隧道围岩的变形是一种危险性强、常见的施工地质灾害,对隧道的安全施工造成了极大的隐患[1]。因此,对隧道围岩的变形监测能实时掌握隧道的变形情况,并根据变形监测数据进行预测分析,对评价围岩的稳定性具有重要的意义[2-4]。同时,随着现代监测技术的快速发展,监测精度较传统监

【相似文献】