风电机组机舱风速多尺度传递机理与前馈控制方法研究

The widely used feedback control for wind turbines has drawback of lag, which often causes delayed yaw alignment, overshoot, and incapability of avoiding gusts. On the contrast, the feedforward control method can overcome these problems, which needs accurate prediction of the incoming wind condition. The normal-used methods of wind speed prediction are mainly for power prediction. These methods are not suitable for feedforward control due to the low time and space accuracy, efficiency and high cost. To solve this problem, a new wind prediction method based on Nacelle Transfer Function of wind speed is proposed. The nacelle anemometer suffers the influence from both the large-scale natural incoming flow and the small-scale surrounding flow of wind turbine, leading to very complicate flow structure in local region. The transfer mechanism between incoming flow and nacelle wind speed is not clear. The wind turbine has no accurate wind speed prediction due to the lack of high accurate transfer model..This project investigates the key scientific problem: the multiscale transfer mechanism of nacelle wind speed. Sturdies will mainly focus on the themes listed as follows. 1) Revealing the mechanisms of turbulence transportation from the incoming flow and the surrounding flow of wind turbine to the micro-scale flow around the nacelle using both numerical simulations wind tunnel experimentss; 2) Building the multi-dimensional numerical simulation model using domain-decomposition strategy, investigating the effect of multiple influential factors, such as, the wind turbine aerodynamic characters and the typical control strategies, on the transfer characters; 3) Building a numerical model of nacelle transfer function with high accuracy based on numerical simulations and experimental results; 4) Investigating the feedforward control strategies of wind turbine under gust, extreme wind conditions, et al, which can be used for the power incensement and load reduction of wind turbine.

风电机组当前采用的反馈控制具有滞后性，容易引起对风不及时、超调、不能有效规避极端风况等问题。采用前馈控制可以克服以上问题，但需要准确预测来流。当前的风速预测主要面向功率预报，其时空精度、测风成本与效率均不能满足前馈控制的要求。针对此问题，本项目提出了“基于机舱传递函数的风速预测方法”。而机舱风速受大尺度湍动来流和小尺度风轮绕流的共同影响，局部流动结构复杂，从来流到机舱风速的传递机理尚不明确，缺乏高精度传递模型。因此，本项目围绕“认识和分析机舱风速多尺度传递机理”这一关键科学问题：1）开展数值模拟和风洞实验，揭示从自然来流、风轮绕流到机舱绕流的湍流输运机理；2）建立多尺度分区计算模型，研究大气条件、典型控制策略等因素对传递特性的影响规律；3）结合实验和数值模拟，建立从来流风速到机舱风速的高精度传递函数模型；4）探索典型风况下风电机组的前馈控制方法，为风电机组提升功率、降低载荷提供支持。

风电机组当前采用的反馈控制具有滞后性，容易引起对风不及时、超调、不能有效规避极端风况等问题。采用前馈控制可以克服以上问题，但需要准确预测来流。当前的风速预测主要面向功率预报，其时空精度、测风成本与效率均不能满足前馈控制的要求。风电机组标配的机舱测风仪为精准的单机风速预测提供了可能。但机舱风速受大尺度湍动来流和小尺度风轮绕流的共同影响，从来流到机舱风速的传递机理尚不明确，缺乏高精度传递模型。针对此问题，本项目提出了“基于机舱传递函数的风速预测方法”，围绕“认识和分析机舱风速多尺度传递机理”这一关键科学问题，开展了数值模拟和风洞实验研究，取得了如下成果：1、首次细致展示了大型风电机组机舱位置近尾流的非定常流动结构，对风电叶片内叶展厚翼型尾流与叶片脱落涡的影响进行了分析，指出进口风速波动对于机舱风速的影响最大，叶片的旋转的影响次之，机舱几何绕流、分离流动等其他的影响十分微小；2、建立了机组上游来流风速与机舱典型位置风速的传递函数，揭示了风轮转速控制、偏航控制与变桨控制对机舱风速的影响规律及修正方法，发现了机舱上方适宜安装风速仪的最优位置及其定位方法；3、研究了基于机舱传递特性的大气湍动来流的风速预测方法，提出了合适的预测时间尺度与滤波器，经过测试，所提出的方法可以较好地辨识机组上游来流的大尺度波动，相比原来基于超短期功率预测的风电机组控制的时间尺度下降了一个量级，由原来的十五分钟量级降至分钟级，可以有效实现极端风况的预测，降低风电机组的极端载荷。同时可以提升风电机组的对风精度，提升发电量。4、进一步研究了漂浮式风电机组在平台多自由度下的动态特性，为下一步进行海上漂浮式风电机组的机舱近尾流特性研究奠定基础。本项目发表论文31篇，其中SCI论文6篇，EI论文12篇，中文核心6篇，申请专利2项，出版译著一本。本项目的研究成果可以用于风电场的智慧运维，提升风电机组发电量的同时，降低机械载荷。