Design of simulation experiment for buried pipeline damage identification based on fiber optic vibration signal
Pipeline plays an important role in urban water supply and drainage,oil and gas transmission and so on.Real-time monitoring of pipeline is very important.In this paper,a deep learning residual network is proposed to classify the optical fiber vibration signals collected by distributed optical fiber vibration sensing system.The optical fiber was laid on the pipeline,and the signals collected by the distributed optical fiber vibration sensing system were regarded as one-dimensional time series.The Grami Angle field was used to encode the one-dimensional time series to obtain the corresponding two-dimensional images.The two-dimensional images were input into the deep learning residual network for classification,so as to realize the real-time monitoring of the pipeline.Experimental results show that this method has higher classification accuracy than converting one-dimensional data into gray image and then input it into deep learning neural network.
distributed fiber optic vibration sensingdeep learningresidual networkGrami Angle field
万方数据
维普
