臺灣漢寶濕地之紊流風場統計特性與模型

Statistical Properties and Model of Turbulent Wind Field in the Near Coastal Area of Han-bao Wetland, Taiwan

李信憲(Hsin-Hsien Li)

指導教授 ： 張建成

共同指導教授 ： 郭志禹

國立臺灣大學/工學院/應用力學研究所/碩士(2015年)

https://doi.org/10.6342/NTU.2015.01828

摘要

本研究將針對永傳在漢寶濕地架設之測風塔所測得之真實資料(real data)進行分析，資料涵蓋2014年12月整個月的量測數據，最後從分析結果進一步地瞭解當地之風況變化與特性。此次風況分析結果中，主要呈現風速隨時間序列之分佈(wind speed time series)、平均風速(mean wind speed)、指數分佈(power law distribution)、韋伯機率分佈(Weibull probability distribution)、風花玫瑰圖(wind rose diagram)、魚骨圖(羽毛圖)、頻譜分析(spectrum analysis)。此外，亦有中央氣象局在芳苑、福興、埔鹽設置測站所測得之實測數據，資料期間為2013年1月至2014年12月。
在頻譜分析當中，由於測風塔與測站之採樣頻率屬低頻範圍，其低頻功率譜密度並不符合Kaimal譜密度模型之趨勢，在此使用Shirer及Rossi (1981)提出適用較低頻之譜密度模型。此外，為看出在高頻範圍的風譜能量是否符合Kaimal功率譜密度的模型，因此，使用採樣頻率較高之LiDAR量測大氣風速，以便觀察較高頻範圍之風譜能量。
另一方面，利用美國國家再生能源實驗室NREL開放授權的紊流產生器TurbSim來模擬一時序列之全場紊流風場，其模擬方法為Sandia法，其概念在於利用空間隔點間的相關性(coherence)，建立頻率域之速度譜(velocity spectra)，最後經由快速傅立葉反轉換(IFFT)運算，求得一時序列之速度分佈。

關鍵字

測風塔 ； 指數分佈 ； 韋伯機率分佈 ； 功率譜密度

並列摘要

This study is about the real data analysis of wind speed measured by wind tower located at Han-bao wetland, from the final results, then, we can realize the local wind characteristics during December 2014. The results will include wind speed time series, mean wind speed, power law distribution, Weibull probability distribution, wind rose diagram, stick plot of wind direction time series, and spectrum analysis. Besides, there are also three groups of data measured by three different anemometers located at Fangyuan, Fuxing, and Puyen, respectively, where the period of these data are all from January 2013 to December 2014.
In spectrum analysis, since the sampling frequency of observation values from wind tower and anemometer are in low-frequency range, the low-frequency power spectral density (PSD) curve doesn’t fit the Kaimal spectrum model. Therefore, another model developed by Shirer and Rossi (1981) will be introduced in this thesis such that the lower-frequency spectrum model could be applicable for real data. To see whether the higher-frequency part would fit the Kaimal spectrum model, use LiDAR to measure the wind speed, i.e., these measured values are higher sampling frequency.
Also, use TurbSim which is a turbulence simulator developed by NREL (National Renewable Energy Laboratory U.S.A) to simulate a full-field time series of turbulence wind farm. The simulation method of TurbSim is well known as Sandia method, its main concept is calculating the coherence of grid points in space, then, constructing the velocity spectra in frequency domain, and finally performing the IFFT to get wind speed.

並列關鍵字

wind tower ； power law distribution ； Weibull probability distribution ； PSD

參考文獻

[1] Arfa Maqsood and Mohammad Aslam, 2008, “A comparative study to estimate the parameters of mixed-Weibull distribution”, Pak.j.stat.oper.res, Vol. IV No.1 2008 pp1-8.

[3] Benedikt Ernst * and J¨org R. Seume, 2012, “Investigation of Site-Specific Wind Field Parameters and Their Effect on Loads of Offshore Wind Turbines”, Energies, ISSN 1996-1997.

[5] D.E. Newland, 1975, “An Introduction to Random Vibrations and Spectral Analysis”, Pages 1-51.

[6] Dieudonné Kidmo Kaoga, Danwe Raidandi, Noël Djongyang, Serge Yamigno Doka, 2014, “Comparison of Five numerical Methods For Estimating Weibull Parameters for Wind Energy Applications in the District of Kousseri, Cameroon”, Asian Journal of Natural & Applied Sciences, Vol. 3(1) March.

[7] D. Winkelaar, 1992, “Program for Three-Dimensional Wind Simulation”, ECN-R--92-013.