H型垂直轴风机非对称尾流及其对风能转换影响机理研究

Vertical axis wind turbines have low cost of energy and a great application potential expanding to the offshore waters. Considering the fact that energy difference upstream and downstream of the turbine determines the energy conversion coefficient, the key index of wind turbine design, it is thus of significant importance to investigate the wake flow precisely and thoroughly. In light of above point, this project will focus on the wake flow structure of the vertical axis wind turbine and its influence on the wind energy conversion. Firstly, the particle image velocimetry will be utilized to obtain the wake flow of a H-type vertical axis wind turbine, the dynamic stall induced flow and the asymmetric wake flow behind the turbine will be obtained via phase-locked technique. Secondly, a precise computational fluid mechanics method will be developed using open-source software OpenFOAM, taking into account of the turbine rotation effect for the flow governing equations and the turbulence model, to simulate the unsteady flow field. The numerical method will be thoroughly verified and validated based on the experimental results. Lastly, numerical studies will be performed to obtain the transient flow around the turbine, then the characteristic flows can be extracted through dynamic mode decomposition method and corresponding energy distribution in the wake and corresponding characteristic flows will be obtained through energy flux method; through parametric analysis and comparison, the mechanism on how the asymmetric wake flow influences the energy conversion will be addressed. The outcomes of this project will provide theoretical basis to assess wind energy conversion according to wake flow information, and also this project is expected to establish a strategy for wake flow analysis by extracting and analyzing characteristic flows based on transient flow information for wind turbines.

垂直轴风机能源成本较低且具有延伸至近海的发展潜力。风机上游和尾流之间能量差决定了关键参数风能转换率，因此尾流分析对风机性能的提高有重要意义。本项目拟结合试验和数值方法，对垂直轴风机尾流及其对风能转换影响开展研究。首先，以H型垂直轴风机模型为对象进行流场粒子图像测速试验，研究动力失速效应并获取和分析非对称尾流的锁相流场。其次，在流动控制方程和湍流模型中考虑风机旋转效应，基于开源程序OpenFOAM建立非稳态风机全流场数值研究方法，结合试验数据对其进行确认和验证。最后，基于数值方法得到的全流场瞬态数据，使用动力模态分解法提取尾流流动特征，利用能量通量法计算尾流及其各特征流动的能量分布并进行参数分析，结合能量转换率变化揭示垂直轴风机非对称尾流对风能转换的影响机理。本项目将为开展基于流场信息的风机风能利用分析提供理论基础，并建立一套基于瞬态流场信息的风机尾流特征提取方法和分析策略。

海上可再生能源工程是国家构建新能源电力系统、实现“碳达峰、碳中和”战略目标的关键组成部分。复杂海洋（风、浪、流等）环境下的海工结构物和装备的设计优化、运维期间状态安全评估等需要流体力学分析作为科学支撑，其中海工结构物尾流诊断与分析是重要的研究手段，如风机风能转换效率、水下和水面船航装备存降噪减阻需求等与结构物尾流状态相关。另一方面，近年来机器学习方法被广泛运用在流动诊断和分析中，如何使机器学习方法契合并服务于海工结构物设计和智能化运行维护具有重要的科学价值和工程意义。.本研究首先基于无监督机器学习动力模态分解（DMD）方法，以层流状态下并排双圆柱交变与偏向振荡尾流的特征流动发展和最终稳定状态为例，探索了DMD方法在揭示特征流动机制方面的应用；随后对湍流状态下垂直轴风机转子内部和外部流动进行了特征流动的动力模态提取和学习，对特征流动的能量分析及受叶尖速比等因素的影响等做出分析；最后，基于有监督的物理信息神经网络PINN方法，研究了机器学习在层流和湍流流动诊断中的应用。主要结论包括：.1. 层流状态下，DMD方法可用于揭示双圆柱交变振荡和偏向尾流的发生机制；.2. 湍流状态下，DMD方法可用于提取垂直轴风机流场特征流动及对应的能量分布；垂直轴风机尾流特征流动呈现非对称性，且主要随叶尖速比而变化；.3. PINN方法可以广泛应用于流动的模拟、重构、逆问题求解等；.4. 本文的案例分析表明，基于参数化Navier-Stokes方程的PINN可应用于湍流流动分析中，且精度较高、对具体流动现象敏感性较低。