Sensitivity analysis of a wind rotor blade utilizing a multi-disciplinary tool chain

Abstract

Due to the multi-disciplinary nature in the design of wind turbines, developers and manufacturers face several challenges. One of them is the exchange of model data between the different disciplines, e.g. aerodynamic or structural dynamics. To overcome the problems of data exchange between the disciplines the Common Parametric Aircraft Configuration Schema (CPACS) has been developed by the German Aerospace Center (DLR). Based on this data scheme a multi-disciplinary tool chain and the verification of their components are presented. Due to some similarities in the design of aircraft wings and wind turbine blades, the tool DELiS has been extended to create structural finite element models of wind turbine blades. The tool has been validated to an industrial rotor and the DTU 10-MW reference rotor blade. The finite element models are dynamically reduced and exported as flexible bodies to the multi-body simulation tool Simpack. For aeroelastic analyses an aerodynamic model is coupled to the multi-body simulation tool. The aerodynamic models can be either a high-fidelity CFD-simulation or a low-fidelity model based blade element momentum theory. This paper illustrates guidelines for the development of a multi-disciplinary tool chain and its interfaces. Based on this framework, sensitivities of parameter changes as well as parameter optimization can be done utilizing a trained neuronal network. The methodology of this analysis with a low number of parameters will allow a sensitivity analysis for complete rotor blade designs in the future.