Analysis of Wake Induced Dynamic Stall on Wind Turbine Blades

本文将研究由上游涡轮尾迹或大气边界层波动引起的非定常入流条件下风力发电机叶片的动态失速现象。在项目的第一阶段，对风力机尾迹进行了分析，重点分析了尖涡和轮毂涡。尾迹是基于大涡模拟预测的。在风电场装置中，采用主动控制来激励紊流尾流中的不稳定模态，以最小化涡轮叶片的动负荷。风力发电机叶片非定常负荷的主要来源是流动分离，在非定常入流条件下，由于上游安装的涡轮机产生的尾迹或大气湍流而经常发生流动分离。也就是说，要将动态失速模型扩展到扰动入流条件和旋转构型，需要风力机叶片动态失速的详细数据。本提案的目的是提高对典型风力发电机应用中动态失速现象的理解和建模。此外，还分析了动态失速对尾流湍流的影响。这些知识将使更准确地预测结构载荷和风力涡轮机的功率输出成为可能。为了准确捕捉边界层生长、分离、向湍流过渡和流动再附着等流动特征，本文将采用基于结构拟合体和非结构笛卡尔混合网格的大涡模拟研究涡轮叶片的动态失速。动态模态分解将用于分析在失速和尾迹动力学中出现的各个模态。这些模拟补充了达姆施塔特工业大学和奥尔登堡大学进行的测量，通过自适应弧度减少动态失速效应，也适用于扫描条件。此外，将对柏林工业大学的风力涡轮机模型进行仿真。

Dynamic stall phenomena on wind turbine blades generated by unsteady inflow conditions from upstream turbine wakes or atmospheric boundary layer fluctuations will be investigated. In the first project phase, wind turbine wakes were analyzed with the focus on the tip and hub vortices. The wake was predicted based on large-eddy simulations. Active control has been applied to excite unstable modes in the turbulent wake to minimize dynamic loads on turbine blades in wind farm installations. A major source of unsteady loads on wind turbine blades is flow separation which frequently happens under unsteady inflow conditions caused by wakes generated by upstream installed turbines or atmospheric turbulence. That is, detailed data of dynamic stall on wind turbine blades are necessary to extend dynamic stall models to perturbed inflow conditions and rotating configurations. The objective of this proposal is to improve the understanding and modeling of dynamic stall phenomena for typical wind turbine applications. In addition, the influence of dynamic stall on the wake turbulence will be analyzed. This knowledge will enable more accurate predictions of structural loads and the power output of a wind turbine. The dynamic stall on the turbine blade will be investigated by large-eddy simulations based on hybrid structured body-fitted and unstructured Cartesian meshes, such that the flow features including boundary layer growth, separation, transition to turbulence and flow reattachment can be captured accurately. Dynamic mode decomposition will be used to analyze the individual modes occurring at stall and in the wake dynamics. These simulations complement the measurements conducted by the TU Darmstadt and Univ. Oldenburg to reduce dynamic stall effects by an adaptive camber also for sweep conditions. Furthermore, simulations for the wind turbine model from the TU Berlin will be performed.