Translated Title
Recognition Method of Pipeline Leakage Acoustic Signals Based on BP Neural Network
Translated Abstract
In view of the urban water supply pipeline leak detection requirements, studied the recognition method of the leakage acoustic signal.Analyzed the time domain, frequency domain and waveform features of the leakage signal, extracted 20 kinds of features can be used to characterize the leakage signal, and build the BP neural network recognition system of the leakage acoustic signal.Researched the neural network structure (the number of hidden nodes, transfer function, learning rate) and the number and variety of input parameters on the leakage signal recognition performance and optimized the best neural network structure and input parameters.Based on the above research, the use of neural networks optimized for laboratory and field pipeline leakage signals cross training and recognition, the results show that the proposed neural network system based on leakage features with high reliability and universality, can well realize the leakage signal under different scenarios of cross recognition, the overall recognition rate of 92.5%.Research work under different working conditions, for solve leak signals identification made beneficial exploration.
Translated Keyword
pipeline leakage
acoustic emission
feature extraction
BP neural network
signal identification
Corresponding authors email
