Modelling and testing of a load-limiting sandwich structure

A heavy gust causes a severe load peak to a Wind Turbine (WT) blade. Such load peaks and the consequent vibrations cause fatigue that reduce the lifetime of the blade. The fatigue loads are also a limiting factor in increasing size of WT blades. This paper deals with a load-limiting structure that will not allow the load peak to grow over a chosen limit. The targeted structure is an adaptive airfoil that automatically reduces camber at a heavy gust, which in turn immediately reduces the loads to the blade. This adaptive concept is totally passive, it needs no heating nor control system. The concept is based on an adaptive laminate that contains superelastic Shape Memory Alloy (SMA) wires in one direction, glass fibers in the other direction and a flexible special epoxy matrix. The feasibility of the system is tested by a planar sandwich beam under 3-point bending. The structure is modelled using the ABAQUS finite element program and two different SMA material models: the Aurichhio-Taylor model existing in ABAQUS commercial code and the new iRLOOP UMAT (ASCR-VTT). The modelled results are compared to each other and to experimental results. The results show that the load-limiting concept is feasible, but the tested structure contains too soft matrix and core. It is found that both models work properly for embedded SMA wires and give nearly identical results, but as they can be used for small strains structural models only they are inadequate for modelling smart structures with large geometric or material nonlinearities outside the SMAs themselves. As smart structures often involve many different nonlinearities it is necessary to modify the SMA UMATs for the finite strains cases.