An Internal Detector Positioning Method in Oil Pipelines Using Vibration Signal

Abstract

In-pipe inspection is an important means to ensure the safety of oil pipelines. External positioning technology of the internal detector is a vital component of the pipeline detection system. Due to the shielding effect of the metal tube wall,a great limit will be exerted on the application of general electromagnetic technology to the tracking and positioning of the internal detector. This paper proposes a positioning method of the internal detector in the pipeline based on the vibration signal. Aiming at the actual operation of the internal detection in the project site, a structural model of the internal detector positioning system based on the vibration signal is designed based on the idea of "coarse classification and exact regression". Through the vibration generated by the friction and collision between the detector and the tube wall, the feature extraction scheme of vibration signal based on wavelet packet decomposition is designed, and the precise positioning of the in-pipe detector is realized by using the two stage positioning network of Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN). Finally, the field experiments verify that the positioning method of internal detector designed in this paper has the advantages of high positioning accuracy and strong anti-interference ability.