On the Importance of Fibre Direction Mesh Alignment for Artificial Lightning Strike Simulations

  • Nagavel, Baskaran (Queen’s University Belfast)
  • Millen, Scott (Queen’s University Belfast)
  • Murphy, Adrian (Queen’s University Belfast)

Please login to view abstract download link

Composite materials, used in primary aircraft structures, produce weight reduction and improved fuel efficiency over legacy metal airframes but are more susceptible to lightning strike damage. Therefore, research into lightning strike damage and protection systems, through experiments and simulations, is an important research topic. For any FE simulation appropriate representation of the material behaviour, the loading and boundary conditions are key to accurate predictions. In addition, an aspect which has been under reported in many studies is the meshing strategy. Fibre direction mesh alignment has been reported to yield more accurate results in the modelling of damage (intralaminar damage initiation and propagation) in unidirectional fibre reinforced composite structures [1,2]. However, this model meshing strategy has not found wide application and has not been used for the modelling of thermal damage events, e.g. lightning strike direct effect simulation. Instead, authors have typically refined the mesh around the arc attachment area [3]. This paper, for the first time, examines the influence of fibre direction mesh alignment for artificial lightning strike simulations and the prediction of thermal damage. The paper uses a mature modelling approach with a transient, fully coupled, thermal-electric step in ABAQUS with a lightning test Waveform A (40 kA, 4/20 µs) applied to the specimen. Specimen boundary conditions match those typically used in experiments and a mesh convergence study is undertaken to ensure no element size influence on the results. The use of this meshing strategy has been shown to significantly improve the prediction of both moderate and severe thermal damage profiles, when compared with the standard meshes used in previous research. The predicted moderate (2659 mm2 vs 2833 mm2) and severe (1059 mm2 vs 1061 mm2) damage areas were improved to within 4% and 1% of experimental results, respectively, using this meshing strategy.