隧道二衬脱空声振检测试验研究

发布时间：2018-08-09 14:45

【摘要】：研究目的:隧道二衬脱空严重威胁线路的安全运营,及时开展检测、预防工作尤为重要。为解决现有检测方法中存在的问题,降低检测成本,本文进行声振检测试验研究,通过对响应信号进行频谱转换、特征值提取、神经网络拟合以得到初步的脱空判定标准。研究结论:(1)严重脱空信号的低频成分(1 000 Hz左右)突出,密实信号的高频成分(8 000 Hz左右)突出,轻微脱空信号表现为比例相近的低、高频成分均匀分布,曲线为多峰值形态;(2)主峰值频率、次峰值频率及峰值下降率可作为信号特征值,全面反映二衬结构状态;(3)将特征值与对应二衬结构状态作为输入、输出值进行BP神经网络训练,并对网络进行测试,预测值的均方误差较小(mse=6.475e-5),网络具有较好的预测能力,能够初步实现对二衬结构状态的定性识别判定;(4)该研究成果可为声振法在隧道二衬脱空检测中的工程运用提供理论依据。
[Abstract]:Research objective: it is very important to carry out inspection and prevention work in time. In order to solve the problems existing in the existing detection methods and reduce the detection cost, the acoustic and vibration detection experiments are carried out in this paper. By means of spectrum conversion, eigenvalue extraction and neural network fitting of the response signal, the preliminary criteria for void determination are obtained. The conclusions are as follows: (1) the low frequency component (about 1 000 Hz) of the serious void signal is prominent, and the high frequency component (about 8 000 Hz) of the dense signal is prominent. (2) the main peak frequency, the secondary peak frequency and the peak drop rate can be regarded as the signal eigenvalues, which can reflect the structure state of the two linings; (3) the eigenvalues and the corresponding two-liner structure states are taken as input and the output values are trained by BP neural network. The test results show that the mean square error (mse=6.475e-5) of the predicted value is small, and the network has better prediction ability. It can preliminarily realize the qualitative identification of the two-liner structure state. (4) the research results can provide the theoretical basis for the engineering application of the acoustic-vibration method in the detection of the two-liner void in the tunnel.
【作者单位】：中铁西北科学研究院有限公司;西安科技大学;
【基金】：中铁科研院科技开发计划项目(2015-KJ027-Z027-03)
【分类号】：U455.91

【相似文献】