Long, Craig SLoveday, Philip WRamatlo, Dineo AMAndhavarapu, EV2018-10-192018-10-192018-09Long, C.S., Loveday, P.W., Ramatlo, D.M. and Andhavarapu, E.V. 2018. Numerical verification of an efficient coupled SAFE-3D FE analysis for guided wave ultrasound excitation. Finite Elements in Analysis and Design, vol 149, pp 45-560168-874X1872-6925https://www.sciencedirect.com/science/article/pii/S0168874X17309472?via%3Dihubhttps://doi.org/10.1016/j.finel.2018.05.001http://hdl.handle.net/10204/10478Copyright: 2018 Elsevier. Due to copyright restrictions, the attached PDF file only contains the abstract of the full text item. For access to the full text item, please consult the publisher's website. The definitive version of the work is published in Finite Elements in Analysis and Design, vol 149, pp 45-56Numerical verification of a method to simulate piezoelectric transducers exciting infinite elastic waveguides is presented. The method, referred to as SAFE-3D, combines a 3D finite element (FE) model of a transducer with a 2D semi-analytical finite element (SAFE) model of the waveguide and accounts for the dynamics of the transducer. An interpolation procedure is employed to transfer forces and displacements between the SAFE and 3D FE models, and therefore nodes at the interface between the two models are not required to be coincident. An Abaqus/Explicit analysis, employing a thermal equivalent piezoelectric model and absorbing boundary conditions to prevent end reflections, is used to verify the accuracy of the SAFE-3D model. A piezoelectric transducer attached to the web of a rail and driven with frequency content which excites a mode cut-off is considered. A driving signal which does not contain cut-off frequencies is used for comparison. Time domain displacement results computed using Abaqus/Explicit and SAFE-3D are compared directly. Several methods to alleviate the numerical difficulties encountered by the SAFE-3D method, when transforming frequency domain displacements to the time domain, close to cut-off frequencies are evaluated. It is shown that post-processing methods have a similar effect to adding damping, but are less numerically expensive if iterative tuning of parameters is required. A SAFE-based method to extract modal amplitudes from Abaqus/Explicit time domain results is used to evaluate the accuracy of SAFE-3D in the frequency domain. Good agreement between the SAFE-3D method and results computed using Abaqus/Explicit is achieved, despite the Abaqus/Explicit and SAFE-3D models predicting slightly different cut-off frequencies.enAbaqus/ExplicitCut-off frequencyModal amplitudePiezoelectric transducer modelSemi-analytical finite elementSAFETrain railArticleLong, C. S., Loveday, P. W., Ramatlo, D. A., & Andhavarapu, E. (2018). Numerical verification of an efficient coupled SAFE-3D FE analysis for guided wave ultrasound excitation. http://hdl.handle.net/10204/10478Long, Craig S, Philip W Loveday, Dineo AM Ramatlo, and EV Andhavarapu "Numerical verification of an efficient coupled SAFE-3D FE analysis for guided wave ultrasound excitation." (2018) http://hdl.handle.net/10204/10478Long CS, Loveday PW, Ramatlo DA, Andhavarapu E. Numerical verification of an efficient coupled SAFE-3D FE analysis for guided wave ultrasound excitation. 2018; http://hdl.handle.net/10204/10478.TY - Article AU - Long, Craig S AU - Loveday, Philip W AU - Ramatlo, Dineo AM AU - Andhavarapu, EV AB - Numerical verification of a method to simulate piezoelectric transducers exciting infinite elastic waveguides is presented. The method, referred to as SAFE-3D, combines a 3D finite element (FE) model of a transducer with a 2D semi-analytical finite element (SAFE) model of the waveguide and accounts for the dynamics of the transducer. An interpolation procedure is employed to transfer forces and displacements between the SAFE and 3D FE models, and therefore nodes at the interface between the two models are not required to be coincident. An Abaqus/Explicit analysis, employing a thermal equivalent piezoelectric model and absorbing boundary conditions to prevent end reflections, is used to verify the accuracy of the SAFE-3D model. A piezoelectric transducer attached to the web of a rail and driven with frequency content which excites a mode cut-off is considered. A driving signal which does not contain cut-off frequencies is used for comparison. Time domain displacement results computed using Abaqus/Explicit and SAFE-3D are compared directly. Several methods to alleviate the numerical difficulties encountered by the SAFE-3D method, when transforming frequency domain displacements to the time domain, close to cut-off frequencies are evaluated. It is shown that post-processing methods have a similar effect to adding damping, but are less numerically expensive if iterative tuning of parameters is required. A SAFE-based method to extract modal amplitudes from Abaqus/Explicit time domain results is used to evaluate the accuracy of SAFE-3D in the frequency domain. Good agreement between the SAFE-3D method and results computed using Abaqus/Explicit is achieved, despite the Abaqus/Explicit and SAFE-3D models predicting slightly different cut-off frequencies. DA - 2018-09 DB - ResearchSpace DP - CSIR KW - Abaqus/Explicit KW - Cut-off frequency KW - Modal amplitude KW - Piezoelectric transducer model KW - Semi-analytical finite element KW - SAFE KW - Train rail LK - https://researchspace.csir.co.za PY - 2018 SM - 0168-874X SM - 1872-6925 T1 - Numerical verification of an efficient coupled SAFE-3D FE analysis for guided wave ultrasound excitation TI - Numerical verification of an efficient coupled SAFE-3D FE analysis for guided wave ultrasound excitation UR - http://hdl.handle.net/10204/10478 ER -