A framework for quick identification of inflow conditions inducing destructive aeroelastic instabilities in wind turbines

Stall-Induced Vibrations (SIV) are an aeroelastic instability that happen when large portions of a wind turbine blade experience an angle of attack such that the slope of the lift coefficient is negative. With a rising necessity to design larger and more flexible wind turbine blades, SIV are an important design consideration as the loads generated during the occurrence might severely damage the blades. The severity of SIV depends on the blade design and inflow conditions.

Exploration of the design and inflow space that cause SIV has a high computational cost, since it involves high fidelity aeroelastic simulations. The aim of this work is to formulate a framework based on an aeroelastic simulator, a surrogate model and an optimizer that can automatically identify the conditions that maximize SIV in a wind turbine with the minimal number of simulations. The framework is based on a type of an optimization algorithm called Surrogate-Based Optimization (SBO), which makes use of surrogate models to approximate an expensive objective functions in the design space and guide the selection of new sample points towards the optimum