Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/21499
標題: 風力發電量隨機模型之研究
Development of a Stochastic Model for Wind Power Generation
作者: 吳予馨
Wu, Yu-Sin
關鍵字: 隨機模型;stochastic model;季節性;最大概似估計;seasonality;maximum likelihood estimation
出版社: 企業管理學系所
引用: 一、中文部分 1. 吳繼平. (2007). 應用類神經網路及基因演算法預測風速與風力發電量. 碩士論文, 中原大學, 桃園縣. 2. 杜逸龍. (2009). 風力發電機發電量之推估. 博士論文, 臺灣大學, 台北市. 3. 徐珮珊. (2008). 股價風險之評估-修正之 Ohlson 股權評價模型與幾何布朗運動之比較. 碩士論文, 中原大學, 桃園縣. 4. 莊月璇. (2001). 台灣地區風速機率分佈之研究. 碩士論文, 國立中央大學, 桃園縣. 5. 莊智強. (2007). 台灣地區風力發電量計算及經濟評估之資料庫系統設計. 碩士論文, 臺灣大學, 台北市. 6. 陳一成. (2007). 台灣風場評估及風力機可用性分析-以台中風力發電廠為例. 碩士論文, 中興大學, 台中市 7. 蔡明志. (2005). 含相乘性跳動之幾何布朗運動 與衍生性商品定價之應用. 碩士論文, 國立成功大學, 台南市. 8. 朱佳仁.(2006). 風工程概論(初版).新北市:科技 9. 康志堅.(2012). 台灣離岸風力發電產業發展趨勢與展望.工研院產業經濟與趨勢研究中心(IEK)報告書 10. 綠色能源產業資訊網http://www.taiwangreenenergy.org.tw/ 11. 方世榮. (2010). 統計學導論(六版).台北市:華泰 12. 楊奕農. (2009). 時間序列分析: 經濟與財務上之應用(二版). 台北市:雙葉 13. 李育嘉. (1985). 漫談布朗運動.數學傳播季刊第9卷第三期.中央研究院數學研究所發行   二、英文部分 1. Cartea, A., & Figueroa, M. G. (2005). Pricing in Electricity Markets: A Mean Reverting Jump Diffusion Model with Seasonality. Applied Mathematical Finance, 12(4), 313-313. 2. Dias,M.,and Rocha,K.(1998)Petroleum concessions with extendible options using mean reversion with jumps to model oil price. Paper presented at the 3rd Annual International Conference on Real Options ,6-8 June, Wassenaar, Leiden, The Netherlands, 3. Dickey, D. A., & Fuller, W. A. (1979). Distribution of the estimators for autoregressive time series with a unit root. Journal of the American statistical association, 74(366a), 427-431. 4. Dixit, A. K., & Pindyck, R. S. (1994). Investment Under Uncertainty. Princeton, New Jersey: Princeton University Press. 5. Global Wind Energy Council(GWEC) (2011) Global Wind Report, Annual market update 2011:Global Wind Energy Council, Belgium 6. Granger, C. W., & Newbold, P. (1974). Spurious regressions in econometrics. Journal of econometrics, 2(2), 111-120. 7. Iacus, S. M. (2008). Simulation and inference for stochastic differential equations: with R examples (Vol. 1): Springer. 8. Mansoor Hamood, A.-H. (2007). Stochastic oil price models:comparison and impact. The Engineering Economist, 52(3), 269-284. 9. Merton, R. C. (1976). Option pricing when underlying stock returns are discontinuous. Journal of Financial Economics, 3(1–2), 125-144.. 10. Morgan, C., Phillips, J., & Jacquemin, J. (2006). Understanding Uncertainties in Energy Production Estimates for Offshore Wind Farms. Paper presented at the Second Conference on World Maritime Technology. 11. Pachamanova, D., & Fabozzi, F. J. (2010). Simulation and Optimization in Finance+ Website: Modeling with MATLAB,@ Risk, or VBA (Vol. 173): Wiley. 12. Phillips, P. C., & Perron, P. (1988). Testing for a unit root in time series regression. Biometrika, 75(2), 335-346. 13. Robert, P., & Daniel, R. (1998). Econometric Models and Economic Forecasts: Irwin and McGraw-Hill, New York. 14. Saphores, J.-D., Khalaf, L., & Pelletier, D. (2002). On jumps and ARCH effects in natural resource prices: An application to Pacific Northwest stumpage prices. American Journal of Agricultural Economics, 84(2), 387-400. 15. Schwartz, E. S. (1997). The stochastic behavior of commodity prices: Implications for valuation and hedging. The Journal of Finance, 52(3), 922-973. 16. Tsay, R. S. (2005). Analysis of financial time series (Vol. 543): Wiley-Interscience. 17. Worldwide Wind EnergyAssociation(WWEA) (2012), Worldwide Wind EnergyAssociation 2012 Half-year report: Worldwide Wind EnergyAssociation,Germany
摘要: 
近年來氣候暖化與能源短缺問題日益嚴重,發展綠色替代能源刻
不容緩。其中,風力發電是目前普遍使用且技術成熟,具商業價值的
再生能源。傳統推估風機發電量的方法是利用台灣歷年風況資料計算
出風速機率分配,並配合標準性能曲線、風機預定高度…等因素來預
估。但即使已有成熟的發電理論公式,由於實際發電量會受到天候變
化或發電機本身遭遇例行性維護或故障維修等因素而有變動,可能影
響預估發電量的準確性。
為了解決實際發電量與預估發電量有不一致的情況,本研究從隨
機程序的觀點,使用「幾何布朗運動」及「均值復歸」兩種隨機程序
建立發電量模型。而針對發電量在某間斷時間點有大幅波動的現象,
我們以「跳躍模型」以及「季節模型」來解釋。之後,針對台灣電力
公司所提供之歷年實際運轉資料進行隨機模式之檢定,並使用最大概
似法估計參數,最後根據估計出的參數進行比較,並藉由模擬發電量
來預測未來的發電量,選擇最適合的風力發電模型。
根據最大概似估計法及模擬值的結果顯示:使用「季節模型」比
使用「跳躍模型」對於發電量波動現象具有較高的解釋能力。而使用
均值復歸模型比使用幾何布朗運動模型更能解釋風力發電量的預估結
果。

In recent years, climate change and energy shortages have been
challenging the world. It is extremely critical to develop green alternative
energies. Among these green energies, wind power is renewable, and has
good performances in terms of energy security, environmental protection
and economic concerns. It is currently competitive and widely used due to
its mature technology and commercial value. The traditional approach to
estimate the energy production of wind turbines is by calculating the
distribution of wind speeds from historical data of the wind regime in
Taiwan, and then combined with the wind turbine height and the nominal
performance curve provided by manufacturers. Even though there are
well-developed theoretical formulas to calculate energy production, but the
actual amount fluctuates due to factors such as weather changes, routine
maintenance, or occasional repair of the generator. These interrelated
uncertainties may affect the accuracy in estimating the energy production.
In order to solve the inconsistency between the actual and the
estimated power generated, this study applies the perspective of stochastic
process to building a wind turbine power model. Two continuous stochastic
processes are in use: Geometric Brownian Motion process (GBM) and
mean reversion process (MR). “Jump model” and “seasonal model” are
used to describe the uncertain factors power plants face, such as fan failure,
turbine maintenance, etc. Afterwards, stochastic process tests are used on
the actual operation data provided by Taiwan Power Company, and then
estimate the parameters with maximum likelihood estimation (MLE).
Finally, the crucial parameters were compared to choose the most suitable
values for wind power generation model.
The result based on maximum likelihood estimation method reveals
that: the “seasonal model” is greater than the “jump process model”, and
MR model has better predictions in the energy production of wind turbines
than GBM model.
URI: http://hdl.handle.net/11455/21499
其他識別: U0005-2706201316184600
Appears in Collections:企業管理學系所

Show full item record
 
TAIR Related Article

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.