Void Detection Technology Of Tunnel Secondary Liner Based On Acoustic-vibration Method

The void of tunnel secondary liner threats the safe operation of road seriously,timely detection and prevention work is particularly significant.In order to solve the problems in existing detection methods,reduce the detection cost,conducting the research of acoustic-vibration detection method.Generalizing the response process of liner under excitation of hammer as the random free vibration problem of local secondary liner block,conducting acoustic-vibration test and numerical modal analysis,based on the analysis and fitting of acoustic signal and modal parameters,putting forward the judgment standard of secondary liner void finally.The main conclusions are as follows:1.The power spectral density function of the response signal is obtained by the frequency spectrum conversion of signal collected through the acoustic-vibration test.Low frequency component（about 1000Hz）of serious void signal is prominent,high frequency component（about 8000Hz）of no void signal is prominent,the slight void signal shows uniform distribution of low and high frequency components in similar proportion,and the curve is in a multi peak shape;the main peak frequency,secondary peak frequency and decline rate of peaks can be regarded as the signal eigenvalues,and reflect the secondary liner structure state in the round.2.Taking the eigenvalues and the corresponding secondary liner structure state as the input and output value for BP neural network training,through testing the network,the mean square error of predictive value is small（mse=6.475e-5）,it means the network has a good prediction ability and can realize the qualitative identification and determination of secondary liner structure state initially.3.For response block under the single excitation of vibration,establishing a typical secondary liner structure model,and doing the numerical modal analysis,the first 5 natural frequencies of the dense block model are obtained,for the void models in different morphological parameters,the main conclusions include:①The fundamental frequency of completely void block is 600-2000Hz,the fundamental values increase with the increase of average void depth in cubic polynomial law,the fitting equation is y = A+B1x+B2x2+B3x3;fundamental values of shallow void block is lower than the deep void block overall;for the same depth,fundamental values of subdued surface void block is higher than inclined surface,the values of level void surface reach the maximum;the first 2 natural frequency differentials in response block is 650-1350Hz,the differentials increase with the increase of average void depth,the change rule is similar to the frequency-average void depth curve.②The fundamental frequency of local void block is 1400-4700Hz,the fundamental values increase with the increase of void length in Logistic function law,the fitting equation is y =A₂+（A₁-A₂）/[1+（x/x0）p];for the same void length,fundamental values of larger initial void depth is higher than the smaller;the first 2 natural frequency differentials in response block is 0-1400Hz,the differentials increase with the increase of void length,the curve has a node at the length of 30cm,it is steeper before the node and the differentials reduce with the increase of initial void depth;the curve is slow after the node and the differentials increase with the increase of initial void depth.③By consulting the corresponding decision graph,the precise morphological parameters of the secondary liner structure can be determined,so the quantitative void identification is realized,then presenting a fast drawing method to obtain the accurate information of the void characteristics.