APPLICATION DE L’ALGORITHME GENETIQUE ET LES ELEMENTS FINIS A L’ETUDE DU COMPORTEMENT A L’ENDOMMAGEMENT DES COMPOSITES SOUMIS A L’IMPACT

Abstract

Composite materials have a wide industrial field of application because of their good mechanical behavior. To limit the cost of mechanical testing, numerical simulation is used by mean of the finite element method and optimization programs based on the genetic algorithm principle. The finite elements dynamic Ls Dyna code is widely used along this thesis. This code has a rich library of different patterns of damage. The rupture model Mat_059 for the laminate and the Mat_020 for the impactor were used. The first step is to ensure that these two models provide satisfactory results. A comparison of the results provided by these models was made with the experimental results according to ASTM D7136. This comparison enabled us to highlight the capacities of these models and to ensure minimal deviations. In order to study the effect of the notch on the impact behavior, a circular notch with a diameter of 5 mm. The composite studied is a cross-laminated T800S / M21 of epoxy matrix reinforced with carbon fibers of sequence [(0/90)6]s. In order to take into account the effect of the geometric shape of the impactor in the presence of circular notch, four impactor were then considered; it is therefore (spherical, truncated cone, hemispherical and cylindrical). Another study focuses on the influence of impact energy. For this, three speeds are used (10m/s, 15m/s and 20m/s). Today, textile preforms exhibit good behavior towards unidirectional composites. The design of the bidirectional and three-dimensional preforms was carried out using the Texgen code developed by the University of Nottingham. For an optimal design of composite subjected to impact loading, the genetic algorithm is well suited. Using the Ls-Optui code interfaced with the Ls Dyna software, the optimum safety factor for a few laminates was determined. The safety coefficient is determined by the Tsai-Wu failure criterion.