A reduced order approach to predict buckling initiated by nonlinear material response in composites

High-performance fiber reinforced plastics are widely spread in aerospace and wind industry. It is challenging to predict the complex damage progression and degradation behavior of composites in case of impact, multiaxial strains and fatigue, their strength and stability.

Rotor blades of wind turbines need to be less susceptible to unexpected failure and require less repair. Technologies as structural health monitoring systems, predictive maintenance concepts or digital twins point in this direction. A prerequisite for the implementation of these techniques are fast analysis methods for these large composite structures in wind industry. For the overall simulation of rotor blades, FE analyses using linear-elastic material models are common today. Nonlinear effects for instance due to plasticity, continuous damage evolution or fatigue have to be investigated on a smaller scale. To go one-step ahead the interaction between nonlinear material response as progressive damage and fatigue and load redistribution as well as buckling phenomena should be taken into account on structure level. This would allow designers to use less conservative procedures, leading to a better structural performance.

Obstacles so far are the excessive computational effort of existing material models and simulation methods. To predict the failure mode and nonlinear response of a neuralgic point (a joint, a ply drop, a stiffness jump etc.), a state-of-the-art detailed finite element analysis would for instance use a layer-based approach incorporating a transversely isotropic elastic-plastic continuum damage model and a cohesive zone model. Such a simulation is only feasible for a small section. Therefore, load redistribution effects and buckling phenomena initiated by nonlinear material response (plasticity, damage, fatigue etc.) cannot be investigated on large structures.

To predict the evolution of static and fatigue damage in composites numerically, an energy approach has been developed at ISD, which allows quantifying the loss of the laminate’s stiffness and strength in a physical manner. Homogenization and sub-modeling techniques were studied. Combining the current versions of ISD’s damage models with an appropriate homogenization technique a new reduced order approach capable to predict buckling initiated by nonlinear material response on structure level shall be developed. The new approach will be calibrated and validated based on experimental results.

• C. Gerendt, A. Dean, T. Mahrholz, N. Englisch, S. Krause, R. Rolfes: On the progressive fatigue failure of mechanical composite joints: Numerical simulation and experimental validation. Composite Structures, 248, 112488, 2020.
  URL: https://www.sciencedirect.com/science/article/abs/pii/S0263822320310552,
  DOI: https://doi.org/10.1016/j.compstruct.2020.112488

• Akterskaia, Margarita, et al. "Progressive delamination analysis through two-way global-local coupling approach preserving energy dissipation for single-mode and mixed-mode loading." Composite Structures 223 (2019): 110892.
  URL: https://www.sciencedirect.com/science/article/abs/pii/S0263822318335049,
  DOI: https://doi.org/10.1016/j.compstruct.2019.110892

Doctoral Researcher: Swami Subramaniyan Venkat 

Scientific Advisors: Raimund Rolfes, Emmanuel Beranger, Olivier Allix