PIEZOELECTRIC SENSOR ARRAY FOR DETECTING GAS PIPELINE CRACK

Abstract

Non-destructive testing (NDT) deals with the inspection of an object for determining its properties without destroying its usefulness. The applications include the detection of cracking in steam generator tubing, nuclear power plants, aircraft and etc. In the oil and gas industry, the traditional pipeline maintenance occurs at defined service intervals, so the structural damages like corrosion and crack occurring in between service intervals are not monitored even though it may be detrimental to the instant health of the pipe’s structure. Other than the pipeline industry, a single method for detecting or monitoring damage in a pipeline does not exist, currently. Instead, industries typically implement a combination of several different techniques that would normally require the pipeline system to be temporarily taken out of operation. Lastly, most of the temporary techniques used are effective in good weather condition and it does not have the probe covering when raining occurs on the site. Considering the limitations of the conventional ultrasonic transducers, guided Lamb Wave offers new opportunities for cases of effective detection of damage in structures. As a method for NDT, the Piezoelectric with Lamb Wave method testing system (PZ-LW) is also used for crack detection in the pipe or plate due to several factors resulting from positioning, depth and shape of crack including the small angle of pipe joining. In this thesis, the development of (PZ-LW) system and combination of optimization and error compensation of piezoelectric sensor array for detection of cracks on the pipeline is used in measuring the defect positioning and width of defect based on-time response and signal feedback amplitude for actual defect measurement. The main idea can be divided into three parts. Firstly, optimize piezoelectric sensor array by using the Design of Expert (DOE) software for Response Surface Methodology (RSM). Secondly, use SimNDT software to simulate the proposed piezoelectric sensor array. The probe design (2D and 3D) model is integrated with the system design and at the same time, the comparison between simulation and PZ-LW technique is executable. Thirdly, to complement this system, the Mamdani Fuzzy Logic is used as an intelligent technique in PZ-LW for high accuracy results. In system development, the graphical interfacing (GUI) is used for graph display on the computer by using MATLAB software and from here, the value of the width of the defect will be identified based on the graph display. Besides, the comparison between the simulation and the actual devices is used for the validation of measurement in this system. The PZ-LW v system design has higher sensitivity and less noise as compared to the conventional technique. The analysis of the experimental results suggests the PZ-LW system to be more accurate in contrast to the previous researcher’s findings by 98.55% of accuracy for the position of defect location. Based on the hole defect, the accuracy achieved to 96.05%. Lastly, the axial defect measurement will exceed 98% of accuracy to identify the shape of the defect. This shows that the merger of the probe and intelligence system which are built will affect the accuracy of the results and it is very useful for the classification of defects apart from the accuracy of the reading displayed. In addition, the comparison of experimental and simulation results in SIMNDT has successfully proven the consistency and accuracy of the proposed PZLW system for carbon steel pipe inspection.