HUNGARIAN JOURNAL OF INDUSTRY AND CHEMISTRY Vol. 51(2) pp. 1–6 (2023) hjic.mk.uni-pannon.hu DOI: 10.33927/hjic-2023-11 STUDY OF COMPOSITE PIPELINES DAMAGED BY CORROSION: CONTROL BY NON-DESTRUCTIVE TESTING SAMIR LECHEB1, ABDELHAKIM DAOUI2, CHAHIRA BRIHMAT3, HAMZA MECHAKRA1, AHMED CHELLIL1, BRAHIM SAFI2*, HOCINE KEBIR4 1 Motor Dynamics and Vibroacoustics Laboratory, Faculty of Technology, M'Hamed Bougara University of Boumerdès, Frantz Fanon City, Boumerdès, 35000, ALGERIA 2 Materials, Processes and Environment Research Unit, Faculty of Technology, M’Hamed Bougara University of Boumerdès, Frantz Fanon City, Boumerdès, 35000, ALGERIA 3 Mechanical Solids and Systems Laboratory, Faculty of Technology, M'Hamed Bougara University of Boumerdès, Frantz Fanon City, Boumerdès, 35000, ALGERIA 4 Roberval Laboratory (UMR 7337), University of Technology of Compiègne, Rue du docteur Schweitzer CS 60319, Compiègne, 60203, FRANCE Since the degradation of material properties and defects often occur in engineering structures due to fatigue loading, it is necessary to develop non-destructive testing methods to assess the safety of engineering structures. In the aerospace industry in particular, the demand for early crack detection to ensure the safety and durability of engineered structures is growing. The automated inspection of pipelines using non-destructive testing (NDT) is desirable because visual inspections are not always consistent. In addition, automated inspections reduce the cost of the inspection process and improve its quality. Since the identification of cracks in engineering materials is very valuable for understanding initial and slight changes in the mechanical properties of materials in complex working environments, numerical simulations of propagating Lamb waves in a cracked pipeline were performed to study their non-linear behavior. Finally, the shapes of all the signal responses can be used to identify the depth, length, shape and orientation of cracks. The elapsed time before signal responses are received varies as a function of the orientation of a crack. Keywords: composite material, pipeline, degradation, damage, non-destructive testing 1. Introduction It is necessary to develop non-destructive testing (NDT) methods to assess the safety of engineering structures. In the aerospace industry in particular, the demand for early crack detection to ensure the safety and durability of engineered structures is growing. Linear ultrasonic testing has been widely used to detect cracks, holes, corrosion and other defects in materials, but is only sensitive to severe defects [1]-[4] through which ultrasonic waves pass [1]-[2]. Therefore, linear ultrasonic testing may well fail to detect closed cracks [2]. Compared to linear ultrasonic testing, long-range and highly sensitive Lamb waves propagate over relatively long distances (a few meters in composites), enabling each ultrasonic pulse to inspect the entire field between the transmitter and receiver, in contrast to traditional step-by-step inspection techniques. The proposed technique will therefore rely on the integration of sensors capable of generating and detecting such waves within or on the surface of the structures to be tested. Such a system Received: 6 May 2023; Revised: 27 May 2023; Accepted: 1 June 2023 *Correspondence: safi_b73@univ-boumerdes.dz must be capable of automatically acquiring, storing and processing data. The presence of a fault will be identified by observing changes in the signal response relative to a reference response recorded before the structure was damaged. Due to the curvature of a tubular structure, the wave properties are more complex than through a plate. Theoretical and numerical analyses of higher-order harmonic generation were conducted on non-linear waveguides of arbitrary cross-sections in weakly non- linear cylinders and plates with large-radius pipes. The simulation showed that cumulative second-harmonic generation with longitudinal, torsional or flexural mode excitation was also observed in pipes when two conditions, namely phase velocity matching and non- zero power flow, as in plates were satisfied. Furthermore, the method of simulating material non-linearity in plates can also be applied to pipes. Experiments concerning material non-linearities in pipes also confirmed the phenomenon of cumulative second-harmonic generation with longitudinal or circumferential wave excitation [4]- [19]. https://doi.org/10.33927/hjic-2023-11 mailto:safi_b73@univ-boumerdes.dz LECHEB, DAOUI, BRIHMAT, MECHAKRA, CHELLIL, SAFI AND KEBIR Hungarian Journal of Industry and Chemistry 2 In this context, the theory and interpretation of the temporal characteristics of Lamb wave signals are mainly based on linear elasticity, i.e. signal characteristics are extracted within the frequency band at which they are generated. In this sense, the temporal characteristics, e.g. the ToF (Timer Off) delay, exhibit to some extent a linear correlation as the material or structural parameters are altered due to damage. Therefore, they are referred to as the temporal features of linear Lamb waves in this study and the associated signal-processing steps are referred to as temporal feature processing. In particular, the ToF delay, one of the simplest but informative linear temporal features, has proven to be effective in locating gross damage, i.e. damage with characteristic dimensions comparable to the wavelength of the sound wave such as open cracks, through holes and voids [5]. 2. Experimental study 2.1. Types of defects in pipelines Regarding the different types of defects, the Pipeline Operators Forum (POF) has classified the different existing defects into various categories [20]. It should be noted that since ultrasonic testing cannot detect cracks that occur perpendicular to a section of pipe, these types of defects are have been disregarded. The following four families of defects are predominant and generally detected during inspections: - delamination; - corrosion; - geometric defects (sinking and ovalization); - metal loss (arc cutting, scratching, grinding, spalling). In the vast majority of cases, these are natural defects, so their characteristics, that is, size, depth, shape, residual texture, etc., vary significantly. The most common forms of pipeline defects are shown in Figures 1a-e. Corrosion is the most frequent initial cause of damage in hydrocarbon pipelines [21] in which the condition of the pipeline deteriorates over time via a singular degradation mechanism. Corrosion degradation can be evaluated in many ways (phenomenologically [22], by making random adjustments [23], using stochastic mechanisms [24], carrying out numerical simulations [25] or conducting empirical studies). Based on these degradation mechanisms, the challenge is to predict how the condition of the pipeline will change between inspections to anticipate any possible internal damage. Given the importance of the fluids transported, containment can result in human, environmental or economic losses. Several factors that have an impact on the evolution of corrosion must be considered, e.g. the temperature in its current state, the initiation of degradation and the chemical composition of the steel. Most publications dealing with ultrasonic scrapers have focused on concrete pipelines [26]. However, the nature of the defects sought in this application is also different as in this type of pipeline only large cracks are dangerous. Finally, many publications have focused on the ultrasonic technology [27]. 2.2. Principle modeling? Leaks and ruptures in pipelines due to ageing and rapid deterioration cost millions of dollars per year, which underlines the necessity of continuous, automatic safety monitoring systems capable of rapidly detecting and warning of defects. Before a major catastrophe occurs, this article examines how sensor networks can detect these problems. Ultrasonic guided waves (Lamb waves) can be employed to transmit energy along the length of the pipeline, making it possible for energy to be transferred wirelessly over a longer distance without being hindered by the electromagnetic shielding of the structure [28]. Research has shown that it is possible to detect defects over a large area using active detection devices for simultaneous actuation and detection. Sensors can be mounted on the surface of the curved pipeline to generate and measure guided waves propagating along the curve [29]. Lamb waves follow the curvature of the structure and detect subsurface defects by measuring the structure on one side and allowing subsurface defects to be detected by measuring the inclination on that side. 2.3. NDT of cracked pipelines Since the application of Lamb waves in the real world is extremely complex, numerical simulations are one of the best ways to understand their behavior and implementing the ABAQUS model is invaluable. This simulation can yield the distribution of displacement and the displacement field, where (see Figure 2): - Geometry: Rin=0.09mm; Rex=0.10mm; - Carbon/epoxy composite material. 3. Results and discussion In this study, two models were used: the first without any Figure 1: The studied types of pipeline defects STUDY OF COMPOSITE PIPELINES 51(2) pp. 1–6 (2023) 3 cracks and the second with. One sensor was created to check the data along the circumference of the model (see Figure 3). This figure also presents the form of the investigated crack. Figure 4 shows that the crack size increases. In the first part of our experiments, we investigated how response signal changes with crack size propagation. 5 sizes of cracks are considered in each model: a1=0.10 mm, a2=0.08 mm, a3=0.06 mm, a4=0.04 mm and a5=0.02 mm. In order to highlight the response behavior, the crack angle  (orientation of the crack) was fixed and the crack size varied as described above as well as shown in Figure 5. The response signals detected during NDT are presented in Figure 6. It was noted that when =0°, crack initiation causes the amplitude of the signal response to decrease. The same is true when =30°, Figure 6 shows that the amplitude decreases as a function of the increase in crack size. When =45°, significant decreases in amplitude following crack initiation can be observed. In the case of =60, the difference between the signals responses is negligible. Logically, when =90°, the signals are identical, which is one of the limitations of the ultrasonic testing of perpendicular cracks. Signal of U2 when =0° Signal of U2 when =30° Signal of U2 when =45° Signal of U2 when =60° Signal of U2 when =90° Figure 6: Response signals detected during NDT when measuring the crack orientation Figure 2: Composite Pipeline Figure 3: NDT of the pipeline Figure 4: Crack formation Figure 5: Crack size propagation LECHEB, DAOUI, BRIHMAT, MECHAKRA, CHELLIL, SAFI AND KEBIR Hungarian Journal of Industry and Chemistry 4 In the second part of our study, with a fixed crack size, crack orientation was changed in each model as follows: =0°, =30°, =45°, =60° and =90° as shown in Figure 7. The results obtained are given in Figure 8 for each crack size studied. This field data shows the importance of global coordinates. Some positive and negative U2 signals at the excitation source can be seen because the global coordinate was located in the y-z plane. It should be noted that the time that elapsed before the signal responses were received varies as a function of the angle of crack orientation. For the purpose of an experimental comparison, the measurements were made in both damaged as well as undamaged areas on flat structures and pipelines. A reference signal was first obtained in the undamaged areas before being received in the damaged ones [28]- [33] The two time signals were superimposed and when the two signals were compared, the signal from an area where a defect is present was amplified, resulting in a frequency shift between the two areas. The characteristics of the Lamb wave time signal were compared using a sensor when different defects were detected in an aluminium tube. Two damage indices were applied based on respective characteristics of Lamb waves. 4. Conclusions It can be concluded that this difference in amplitude is due to the absence of matter (a discontinuity) preventing a part of the waves from propagating through the pipeline walls, resulting in significant attenuation of the signal amplitude of the waves. This phenomenon is referred to as the scattering of Lamb waves. When =0°, crack initiation causes the amplitude of the signal response to attenuate. When =30°, the amplitude decreases as the crack size increases. From the graph depicting =45°, a significant reduction in the amplitude is observed during crack initiation. However, when =60°, the difference between the signals responses is negligible. Logically, when =90°, the signals are identical. This phenomenon, known as perpendicular cracking, is one of the limitations of ultrasonic testing. The signal responses when no cracks are present and when =0o are identical. The time elapsed before signal responses are received varies as a function of the angle of crack orientation. Finally, the shapes of all these signal responses can be used to identify the depth, length, shape and orientation of cracks. Figure 7: Crack propagation Signal of U2 when a=0.10 mm Signal of U2 when a=0.08 mm Signal of U2 when a=0.06 mm Signal of U2 when a=0.04 mm Signal of U2 when a=0.02 mm Figure 8: Response signals detected during NDT when measuring the size of the cracks STUDY OF COMPOSITE PIPELINES 51(2) pp. 1–6 (2023) 5 REFERENCES [1] Barkanov, E.N.; Dumitrescu, A.; Parinov, I.A.: Non-destructive testing and repair of pipelines (Springer International Publishing AG), 2018, DOI: 10.1007/978-3-319-56579-8 [2] Mingxi, D.: Cumulative second-harmonic generation accompanying nonlinear shear horizontal mode propagation in a solid plate, J. Appl. Phys., 1998, 84(7), 3500–3505, DOI: 10.1063/1.368525 [3] Praetzel, R.; Clarke, T.; Schmidt, D.; de Oliveira, H.; Dias da Silva, W. C.: Monitoring the evolution of localized corrosion damage under composite repairs in pipes with guided waves, NDT E Int., 2021, 122, 102477, DOI: 10.1016/j.ndteint.2021.102477 [4] Truong, T.C.; Lee, J.-R.: Thickness reconstruction of nuclear power plant pipes with flow-accelerated corrosion damage using laser ultrasonic wavenumber imaging, Struct. Health Monit., 2018, 17(2), 255–265, DOI: 10.1177/1475921716689733 [5] Sobkiewicz, P.; Bieńkowski, P.; Błażejewski, W.: Microwave non-destructive testing for delamination detection in layered composite pipelines, Sensors, 2021, 21(12), 4168, DOI: 10.3390/s21124168 [6] Haniffa, M.A.M.; Hashim, F.M.: Recent developments in in-line inspection tools (ILI) for deepwater pipeline applications, Proc. 2011 Natl. Postgrad. Conf., 2011, 6136416, DOI: 10.1109/NatPC.2011.6136416 [7] Jones, T.S.; Polansky, D.; Berger, H.: Radiation inspection methods for composites, NDT E Int., 1988, 21(4), 277–282, DOI: 10.1016/0308-9126(88)90341- 0 [8] Clauzon, T.; Thollon, F.; Nicolas, A.: Flaws characterization with pulsed eddy currents NDT, IEEE Trans. Magn., 1999, 35(3), 1873–1876, DOI: 10.1109/20.767399 [9] Murphy, K.; Lowe, D.: Evaluation of a novel microwave based NDT inspection method for polyethylene joints, Proc. ASME 2011 Pressure Vessels and Piping Conf., 2011, 5, 321–327, DOI: 10.1115/PVP2011-58086 [10] Zhu, X.W.; Pan, J.P.; Tan, L.J.: Microwave scan inspection of HDPE piping thermal fusion welds for lack of fusion defect, Appl. Mech. Mater., 2013, 333-335, 1523–1528, DOI: 10.4028/www.scientific.net/AMM.333-335.1523 [11] Carrigan, T.D.; Forrest, B.E.; Andem, H.N.; Gui, K.; Johnson, L.; Hibbert, J.E.; Lennox, B.; Sloan, R: Nondestructive testing of nonmetallic pipelines using microwave reflectometry on an in-line inspection robot, IEEE Trans. Instrum. Meas., 2018, 68(2), 586–594, DOI: 10.1109/TIM.2018.2847780 [12] Bates, N.; Lee, D.; Maier, C.: A review of crack detection in-line inspection case studies, Proc. 2010 8th Int. Pipeline Conf., 2010, 1, 197–208, DOI: 10.1115/IPC2010-31114 [13] Slaughter, M.; Spencer, K.; Dawson, J.; Senf, P.: Comparison of multiple crack detection in-line inspection data to assess crack growth, Proc. 2010 8th Int. Pipeline Conf., 2010, 1, 397–406, DOI: 10.1115/IPC2010-31255 [14] Wang, H.; Yajima, A.; Liang, R.Y.; Castaneda, H.: A Bayesian model framework for calibrating ultrasonic in-line inspection data and estimating actual external corrosion depth in buried pipeline utilizing a clustering technique, Struct. Saf., 2015, 54, 19–31, DOI: 10.1016/j.strusafe.2015.01.003 [15] Varela, F.; Tan, M.Y.; Forsyth, M.: An overview of major methods for inspecting and monitoring external corrosion of on-shore transportation pipelines, Corros. Eng. Sci. Technol., 2015, 50(3), 226–235, DOI: 10.1179/1743278215Y.0000000013 [16] Tse, P.W.; Mathew, J.; Wong, K.; Lam, R.; Ko, C.N. (Eds.): Engineering asset management - systems, professional practices and certification (Springer International Publishing Switzerland), 2015, DOI: 10.1007/978-3-319-09507-3 [17] Shafeek, H.I.; Gadelmawla, E.S.; Abdel-Shafy, A.A.; Elewa, I.M.: Automatic inspection of gas pipeline welding defects using an expert vision system, NDT E Int., 2004, 37(4), 301–307, DOI: 10.1016/j.ndteint.2003.10.004 [18] Guan, R.; Lu, Y.; Wang, K.; Su, Z.: Fatigue crack detection in pipes with multiple mode nonlinear guided waves, Struct. Health Monit., 2019, 18(1), 180–192, DOI: 10.1177/1475921718791134 [19] Hong, M.; Su, Z.; Lu, Y.; Sohn, H.; Qing, X.: Locating fatigue damage using temporal signal features of nonlinear Lamb waves, Mech. Syst. Signal Process., 2015, 60-61, 182–197, DOI: 10.1016/j.ymssp.2015.01.020 [20] Pipeline Operators Forum: Specifications and requirements for intelligent pig inspection of pipelines (Version 2009) [21] Performance of European cross-country oil pipelines - Statistical summary of reported spillages in 2012 and since 1971 (Report) CONCAWE, 2013 [22] Tang, P.; Yang, J.; Zheng, J.; Wong, I.; He, S.; Ye, J.; Ou, G.: Failure analysis and prediction of pipes due to the interaction between multiphase flow and structure, Eng. Fail. Anal., 2009, 16(5), 1749–1756, DOI: 10.1016/j.engfailanal.2009.01.002 [23] Amaya-Gómez, R.; Sánchez-Silva, M.; Munoz, F.: Pattern recognition techniques implementation on data from In-Line Inspection (ILI), J. Loss Prev. Process Ind., 2016, 44, 735–747, DOI: 10.1016/j.jlp.2016.07.020 [24] Zhang, S.; Zhou, W.: Cost-based optimal maintenance decisions for corroding natural gas pipelines based on stochastic degradation models, Eng. Struct., 2014, 74, 74–85, DOI: 10.1016/j.engstruct.2014.05.018 [25] Li, S.-X.; Yu, S.-R.; Zeng, H.-L.; Li, J.-H.; Liang, R.: Predicting corrosion remaining life of underground pipelines with a mechanically-based probabilistic model, J. Pet. Sci. Eng., 2009, 65(3-4), 162–166, DOI: 10.1016/j.petrol.2008.12.023 https://doi.org/10.1007/978-3-319-56579-8 https://doi.org/10.1007/978-3-319-56579-8 https://doi.org/10.1063/1.368525 https://doi.org/10.1063/1.368525 https://doi.org/10.1016/j.ndteint.2021.102477 https://doi.org/10.1177/1475921716689733 https://doi.org/10.3390/s21124168 https://doi.org/10.1109/NatPC.2011.6136416 https://doi.org/10.1109/NatPC.2011.6136416 https://doi.org/10.1016/0308-9126(88)90341-0 https://doi.org/10.1016/0308-9126(88)90341-0 https://doi.org/10.1109/20.767399 https://doi.org/10.1109/20.767399 https://doi.org/10.1115/PVP2011-58086 https://doi.org/10.1115/PVP2011-58086 https://doi.org/10.4028/www.scientific.net/AMM.333-335.1523 https://doi.org/10.4028/www.scientific.net/AMM.333-335.1523 https://doi.org/10.1109/TIM.2018.2847780 https://doi.org/10.1115/IPC2010-31114 https://doi.org/10.1115/IPC2010-31114 https://doi.org/10.1115/IPC2010-31255 https://doi.org/10.1115/IPC2010-31255 https://doi.org/10.1016/j.strusafe.2015.01.003 https://doi.org/10.1179/1743278215Y.0000000013 https://doi.org/10.1007/978-3-319-09507-3 https://doi.org/10.1016/j.ndteint.2003.10.004 https://doi.org/10.1016/j.ndteint.2003.10.004 https://doi.org/10.1177/1475921718791134 https://doi.org/10.1016/j.ymssp.2015.01.020 https://doi.org/10.1016/j.ymssp.2015.01.020 https://doi.org/10.1016/j.engfailanal.2009.01.002 https://doi.org/10.1016/j.jlp.2016.07.020 https://doi.org/10.1016/j.jlp.2016.07.020 https://doi.org/10.1016/j.engstruct.2014.05.018 https://doi.org/10.1016/j.engstruct.2014.05.018 https://doi.org/10.1016/j.petrol.2008.12.023 LECHEB, DAOUI, BRIHMAT, MECHAKRA, CHELLIL, SAFI AND KEBIR Hungarian Journal of Industry and Chemistry 6 [26] Iyer, S.; Sinha, S.K.; Tittmann, B.R.; Pedrick, M.K.: Ultrasonic signal processing methods for detection of defects in concrete pipes, Autom.Constr., 2012, 22, 135–148, DOI: 10.1016/j.autcon.2011.06.012 [27] Kim, H.W.; Lee, J.K.; Kim, Y.Y.: Circumferential phased array of shear-horizontal wave magnetostrictive patch transducers for pipe inspection, Ultrasonics, 2013, 53(2), 423–431, DOI: 10.1016/j.ultras.2012.07.010 [28] Tseng, V.F.-G.; Bedair, S.S.; Radice, J.J., Drummond, T.E.; Lazarus, N.: Ultrasonic Lamb waves for wireless power transfer, IEEE Trans. Ultrason. Ferroelectr. Freq. Control, 2020, 67(3), 664–670, DOI: 10.1109/TUFFC.2019.2949467 [29] Jin, Y.; Eydgahi, A.: Monitoring of distributed pipeline systems by wireless sensor networks, Proc. 2008 IAJC-IJME Int. Conf, 2008, Paper 213, IT 304, https://ijme.us/cd_08/PDF/213%20%20IT304.pdf [30] Wang, X.; Tse, P.W.; Mechefske, C.K.; Hua, M.: Experimental investigation of reflection in guided wave-based inspection for the characterization of pipeline defects, NDT E Int., 2010, 43(4), 365–374, DOI: 10.1016/j.ndteint.2010.01.002 [31] Demma, A.; Cawley, P.; Lowe, M.; Roosenbrand, A.G.; Pavlakovic, B.: The reflection of guided waves from notches in pipes: A guide for interpreting corrosion measurements, NDT E Int., 2004, 37(3), 167–180, DOI: 10.1016/j.ndteint.2003.09.004 [32] Füvesi, B.; Ulbert, Z.: GPU-accelerated simulation of a rotary valve by the discrete element method, Hung. J. Ind. Chem., 2019, 47(2), 31–42, DOI: 10.33927/hjic-2019-18 [33] Gergely, A.; Kristóf, T.: Corrosion protection with ultrathin graphene coatings: a review, Hung. J. Ind. Chem., 2021, 41(2), 83–108, DOI: 10.1515/508 https://doi.org/10.1016/j.autcon.2011.06.012 https://doi.org/10.1016/j.ultras.2012.07.010 https://doi.org/10.1016/j.ultras.2012.07.010 https://doi.org/10.1109/TUFFC.2019.2949467 https://ijme.us/cd_08/PDF/213%20%20IT304.pdf https://doi.org/10.1016/j.ndteint.2010.01.002 https://doi.org/10.1016/j.ndteint.2003.09.004 https://doi.org/10.33927/hjic-2019-18 https://doi.org/10.33927/hjic-2019-18