應用中尺度到微尺度極端風模型於風機之負載分析

Abstract

本文旨在利用先進的風場模型探討臺灣離岸風機於極端風速下之負載，並利用一歷史之颱風事件為案例進行分析，計算風機受到真實極陣風以及處於紊流下時產生的結構響應，並進行後續的延伸評估，包含考慮該風機是否同時受到附近其他風機的尾流效應影響，以及使用大量不同風場及狀況下的風機模擬結果進行一系列統計分析。在以往的極端風場模擬中，經常遇到無法同時考量影響颱風事件的中尺度大氣參數及微尺度的複雜紊流結構的問題，導致模擬結果與實際結果有所出入。為了解決這個問題，本研究針對模擬流程及方法進行了改良。研究中將通過天氣研究與預報模式 (weather research and forecasting, WRF) 模擬獲得的颱風風速、溫度及壓力等資訊，以邊界條件的形式導入一解析度較高的大渦流模擬 (large-eddy simulation, LES) 中，微尺度大渦流模擬使用美國國家再生能源實驗室NREL (National Renewable Energy Laboratory)所開發可應用於大氣邊界層紊流分析的SOWFA程式(第6版)，通過模擬計算獲得在時間上以及在空間上更高解析度的風場。取得高解析度風場後，再以致動線模型計算風速轉換至風機上的受力以及尾流，研究中使用了兩種不同軟體，分別是開放軟體SOWFA與OpenFAST，利用這兩個軟體計算並評估風機於尾流影響以及不同非運作狀況下受到的各種負載，負載種類包含葉片根部出轉子平面向彎矩(OoPBM)、塔底前後向彎矩(FATBM)以及機艙水平擺動扭矩(TTYM)三種。模擬中使用到的風機模型是採用NREL所開發的NREL-5MW以及參考IEA規範設計的IEA-15MW-240-RWT，並以這兩種風機模型做為風力發電場內兩種研究案例之目標風機。於風機尾流模擬案例中，結果顯示因尾流影響上游風機之輪轂高度風速為21.9 m/s、下游風機之輪轂高度風速為21.3 m/s，上游風速較下游風速大，但葉片負載則為下游風機38.75 MN-m、上游風機29.99MN-m，下游負載較大。於不同風機非運作狀態下負載統計分析案例中，相同條件下普通停機(parked)與無負荷(idling)葉片根部彎矩負載值相差2.97 MN-m，三種平擺角度(0、45、90度)葉片根部值分別為5.39、4.49、1.76 MN-m。結果顯示不當的停機狀態對風機負載也有極大的影響。本研究期許能夠建立一套完善且可靠的分析流程，足以做為未來評估臺灣離岸風電場受颱風侵襲時，在極端負載下準確的設計依據。

This paper studies the offshore wind turbine loads under extreme wind speeds in Taiwan using advanced wind field modelling. Based on a historical typhoon event, the structural response of turbines under realistic extreme gusts and turbulences is investigated. Subsequent extended assessments are also conducted, including the consideration of wake effects from nearby turbines and a series of statistical analyses using simulation results under wind turbines in different wind conditions. Traditional models for extreme wind simulation usually fail to include both the mesoscale atmospheric parameters, which are essential to typhoon events, and the complex structures of turbulence in microscales. This discrepancy often leads to disparities between the simulation results and the actual observations. To address this issue, this study has made improvements to the simulation process and methodology. In this study, we use the Weather Research and Forecasting (WRF) mode to simulate the typhoon fields and feed the derived data of wind velocities, temperatures and pressures, as boundary conditions, into a higher resolution model of large-eddy simulation (LES). The microscale LES employs the SOWFA v6 program developed by the National Renewable Energy Laboratory (NREL) for analyzing turbulent flows in the atmospheric boundary layer. Through these simulations, fields with higher temporal and spatial resolutions can be obtained. The fields are then coupled with turbine models through the methods of actuator line. The forces exerted on the turbines and the wakes resulting from the inflow velocities are in turn computed by the open-source software, SOWFA and OpenFAST, which allow to estimate turbine loads with the wake effects and to discuss the loads when turbine is operated in different conditions. Three considered types of loads include the out-of-plane blade root bending moment (OoPBM), the fore-aft tower base bending moment (FATBM), and the yawing moment of the nacelle (TTYM). As for the wind turbine model used in the simulations, we take the NREL-5MW and IEA-15MW-240-RWT developed by NREL, which serves as the target turbine for two different studies within the wind farm for offshore wind power generation. In the results of Case 1, the hub-height wind speed for the upstream turbine affected by wake is 21.9 m/s, while for the downstream turbine, it is 21.3 m/s. The wind speed is higher upstream compared to downstream. However, the blade load for the downstream turbine is 38.75 MN-m, while for the upstream turbine, it is 29.99 MN-m. The loads are higher for the downstream turbine compared to the upstream turbine. In the results of Case 2, under the same conditions, the OoPBM values differ by 2.97 MN-m between normal parked and idling modes. The OoPBM values for three yaw angles (0, 45, 90 degrees) are 5.39 MN-m, 4.49 MN-m, and 1.76 MN-m, respectively. The results indicate that improper parked conditions also have a significant impact on the turbine loads. This research project aims to establish a complete and solid framework of advanced simulation processes for future estimation of the extreme loads on offshore wind farms in Taiwan during the impact of typhoons.