Open AccessPipeline Monitoring using Ultrasonic Sensors
Author(s) -
K Raghavendra,
Swetha,
Svk Naresh
Publication year - 2019
Publication title -
international journal of engineering and advanced technology
Language(s) - English
Resource type - Journals
ISSN - 2249-8958
DOI - 10.35940/ijeat.f1373.0986s319
Subject(s) - pipeline transport , nondestructive testing , signal (programming language) , ultrasonic sensor , acoustics , ultrasonic testing , pipeline (software) , reliability (semiconductor) , transducer , wavelet , calibration , fast fourier transform , condition monitoring , computer science , engineering , mechanical engineering , artificial intelligence , electrical engineering , medicine , power (physics) , statistics , physics , mathematics , algorithm , quantum mechanics , radiology , programming language
Pipelines used to transport oil and gas deteriorate with time due to various reasons including corrosion, deformation, cracking etc., So instead of critical-event based inspections, a means of continuous monitoring of pipelines to analyze its response as a function of time is required. Using real time data, this monitoring can be used as a means of prior warning of any unsafe condition of many types of nondestructive testing (NDT) techniques used for defect detection in pipelines, Ultrasonic testing (UT) is preferred because of its high accuracy, reliability, affordability and its capability of generating instant test report. UT works on the principle of pulse-echo method. As the received echo signal by transducer is noisy, it is generally processed using wavelet transforms and/or FFT. Later the signal is given to Artificial Neural Network (ANN) to distinguish signal with Non-Defect (ND) and signal with Defect (D). In this proposal, UT technique is used in pipeline monitoring during machining and real time work. Two calibration standards such as straight beam testing and angle beam testing are used to detect parallel defects and weld inspection respectively.