THE ANALYSIS SYSTEM OF UNDERGROUND PIPELINECOATING DETECTION AND ITS APPLICATION

摘要

The quick analysis system for under ground pipeline coating detection without excavation hasbeen set up. The close interval potential survey (CIPS) and the galvanostatic transient response methodwere used to detect the coating state of buried pipelines. The coating damage point and the coating statewere diagnosed with the analysis system on the spot. An algorithm was presented based on the Dyadicwavelet transform for detecting the pipeline coating damage point with the Close Interval Pipe-to-Soilpotential data. A six-layer Wavelet Kohonen Neural Network model was set up with the galvanostatictransient response to diagnose the states of pipeline coating. The front five layers of the network werecorresponding to multi-resolution decompositions, which has the ability of picking up the information.The last layer of the model was corresponding to Kohonen Neural Network, which has the ability ofself-training. The model can diagnose the coating state with the smooth coefficients with the last layer ofthe model. The detection results of real pipelines of northern China are satisfactory.