基于声振法的隧道二衬脱空检测技术研究

发布时间：2018-03-04 10:34

本文选题：声振法　切入点：隧道二衬脱空检测　出处：《西安科技大学》2017年硕士论文　论文类型：学位论文

【摘要】：隧道二衬脱空严重威胁道路的安全运营,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,进行声振法检测技术研究。将锤击激励下的衬砌响应过程概化为局部二衬块体的随机自由振动问题,进行声振检测试验及数值模态分析,通过对声振信号及模态参数的分析拟合,最终提出二衬脱空判定标准。主要结论如下:(1)通过检测试验所采集声振信号的频谱转换,获取响应信号的功率谱密度函数。严重脱空信号的低频成分(1000Hz左右)突出,密实信号的高频成分(8000Hz左右)突出,轻微脱空信号表现为比重相近的低、高频成分均匀分布,曲线为多峰值形态;主峰值频率、次峰值频率及峰值下降率可为作为信号特征值,全面反映二衬结构状态。(2)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,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,可初步实现对二衬结构状态的定性识别判定。(3)针对单次振动激励下的响应块体,建立典型二衬结构模型,进行数值模态分析,得到密实块体模型的前5阶固有频率,对于不同形态参数的脱空模型,主要结论包括:①完全脱空响应块体基频为600-2000Hz,基频值随平均脱空深度增大呈三次多项式规律增大,拟合方程为y=A+B1X+B2X2+B3X3;浅部脱空块体的基频值整体低于较深处脱空块体;平均深度相同时,平缓脱空面较倾斜脱空面块体有更大的基频值,脱空面为水平时基频达到最大值;响应块体前两阶固有频率差值为650-1350Hz,频率差值随平均脱空深度的增大而增大,变化规律同基频——平均脱空深度曲线类似。②局部脱空响应块体基频为1400-4700Hz,基频值随脱空长度的增大呈Logistic函数规律减小,拟合方程为y=A_2+(A_1-A_2)/[1+(x/x0)p];脱空长度相同时,起始脱空深度越大,基频值越大;响应块体前两阶固有频率差值为0-1400Hz,频率差值随脱空长度的增大而增大,曲线在脱空长度为30cm处存在节点,节点之前曲线较陡,起始脱空深度越大,频率差值越小;节点之后曲线较缓,起始脱空深度越大,频率差值越大。③通过查阅对应的判定曲线图,可确定二衬结构的精确形态参数,实现定量脱空判定,并提出了快速求得脱空特征精确信息的作图法。
[Abstract]:In order to solve the problems existing in the existing detection methods and reduce the detection cost, it is particularly important to carry out the inspection and prevent the tunnel second lining clearance seriously, which is a serious threat to the safe operation of the road. The acoustic vibration detection technique is studied. The response process of the lining under hammering excitation is generalized as the random free vibration problem of the local two-liner block, and the acoustic vibration detection test and numerical modal analysis are carried out. By analyzing and fitting the acoustic and vibration signals and modal parameters, the criteria for determining the void of the two linings are put forward. The main conclusions are as follows: 1) the spectrum conversion of the acoustic and vibration signals collected by the detection test. The power spectral density function of the response signal is obtained. The low frequency component of the serious empty signal is about 1000 Hz) and the high frequency component of the dense signal is about 8 000 Hz). The main peak frequency, the secondary peak frequency and the peak drop rate can be regarded as the signal eigenvalues, reflecting the two-lining structure state in an all-round way.) the eigenvalue and the corresponding two-liner structure state are taken as input and output values for BP neural network training, and the main peak frequency, the secondary peak frequency and the peak drop rate can be regarded as the signal eigenvalues. When the network is tested, the mean square error of the predicted value is smaller than 6.475e-5, and the network has better prediction ability. It can preliminarily realize the qualitative identification judgment of the state of the two-liner structure. (3) for the response block under the single vibration excitation, the typical two-lining structure model is established. The first five natural frequencies of the dense block model are obtained by numerical modal analysis, and the void model with different shape parameters is obtained. The main conclusions are as follows: the fundamental frequency of the complete void response block is 600-2000Hz, and the fundamental frequency increases with the increase of the average void depth by cubic polynomial law, and the fitting equation is YYAB1X B2X2B3X3, and the fundamental frequency of the shallow void block is lower than that of the deep void block as a whole. When the average depth is the same, the fundamental frequency of the flat surface is larger than that of the inclined surface block, and the fundamental frequency reaches the maximum when the void surface is horizontal, and the frequency difference increases with the increase of the average void depth, and the difference of the first two natural frequencies of the block is 650-1350Hz. The law of variation is similar to the curve of the basic frequency-average void depth. 2. The basic frequency of the local emptying response block is 1400-4700Hz. the fundamental frequency value decreases with the increase of the emptying length, and the fitting equation is YSP-A2A1-A1-A2T / [1 + x / xx0p]. The greater the initial emptying depth is, the greater the initial emptying depth is when the emptying length is the same. The larger the fundamental frequency is, the larger the frequency difference between the first two steps of the response block is 0-1400Hz, and the frequency difference increases with the increase of the empty length. There are nodes in the curve where the void length is 30cm, the curve before the node is steeper, the greater the initial void depth is, the smaller the frequency difference is. The curve behind the node is slower, the greater the initial void depth is, the greater the frequency difference is. 3. By consulting the corresponding judgment curve, the precise shape parameters of the two-lining structure can be determined, and the quantitative void determination can be realized. A mapping method is proposed to quickly obtain the accurate information of the void feature.
【学位授予单位】：西安科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U455.91

【参考文献】