Adjoint-Based High-Fidelity Structural Optimization of Wind-Turbine Blade for Load Stress Minimization

Abstract

A high-fidelity multidisciplinary modeling and optimization capability is employed for optimization of structural properties of the 13 m Scaled Wind Farm Technology Facility wind-turbine blade of Sandia National Laboratories. This involves coupling a Reynolds-averaged Navier–Stokes fluid dynamics solver with a structural finite element solver. The exact sensitivities of performance objectives are obtained using the discrete adjoint method. For each of several load cases, the composite fiber angles throughout the internal structure of the blade are optimized to minimize a scalar stress parameter that correlates with the accumulation of fatigue damage. Three main types of loads regularly experienced by a wind-turbine blade are identified: aerodynamic loads, centrifugal loads, and gravitational loads. Static structural optimization is performed with the blade under centrifugal and aerodynamic loads, and dynamic optimization with the blade under gravitational loads. Then, static and dynamic optimizations are performed again with all combined loads present. A 18–60% reduction in the driving stress for fatigue is seen in after optimization.