
系統識別號 
U00261507202016205700 
論文名稱(中文) 
利用類神經網路模型進行彰濱海域的短期離岸風速預測 
論文名稱(英文) 
ShortTerm Wind Speed Forecasting Using Neural Network Models for Chanbin Offshore Area 
校院名稱 
成功大學 
系所名稱(中) 
機械工程學系 
系所名稱(英) 
Department of Mechanical Engineering 
學年度 
108 
學期 
2 
出版年 
109 
研究生(中文) 
孫億昇 
研究生(英文) 
YeeSheng Soon 
電子信箱 
n16075023@mail.ncku.edu.tw 
學號 
N16075023 
學位類別 
碩士 
語文別 
英文 
論文頁數 
105頁 
口試委員 
指導教授林大惠 口試委員蔡英聖 口試委員余政達 口試委員蘇威智

中文關鍵字 
風速
小波頻譜
長短期記憶模型
風速預測

英文關鍵字 
Wind energy
Wavelet Spectrum
Long ShortTerm Memory
Wind Forecasting

學科別分類 

中文摘要 
2016年，台灣政府發佈了一個新能源政策，為求期望台灣能在2025年成為一個非核家園。台灣西半部因地勢的關係在秋冬季有著強而穩定的東北季風，因此離岸風力在替補核電廠停運之後的能源缺口上，成為了萬眾矚目的焦點。風力發電最大的問題在於風的不穩定性。一個好的短期預測有利於風機的控制並產生更大的效能，也能減少此不穩定性對電網產生的傷害。本研究將針對台電氣象觀測塔2017至2019年的風速風向觀測資料進行基本的分析，並藉由分析結果了解預測模型的有效性。本次研究中，主要以類神經網路模型的方式進行短時多步的風速預測，並藉由改變模型參數的方式優化模型。
數據分析的部分分為：趨勢分析，統計分析以及頻譜分析三個部分。趨勢分析以時序圖以及風瑰圖帶出三年風速風向的趨勢。統計分析的部分則在於以月份以及季節的統計值探討每年的異同。最後再以小波轉換得出的頻譜了解各個短尺度時間段的不同情況風場的特徵。在模型建立好之後，除了會進行一般的模型評估之外，也會藉由先前的分析結果作進一步的理解預測產生巨大誤差的原因。並給予建議。本次所使用的類神經預測模型為目前較為廣泛使用的LSTM模型。基於對模型演算法的認識有限，本次參數的矯正方向則傾向於非模型架構的部分。力求從實際氣候資料上得到較高的準確值。
最後，本次研究在最後發現，每年春、秋、及冬季的風況大同小異。而夏季則容易受到不定期的西南季風以及颱風的侵襲影響。秋冬的風速平均較高 (~12m/s)，風向也很穩定（~18°）；春夏的風速平均較低 (~7m/s)且風向的位置較不特定。小波轉換得出的結果發現風況大致可以用東北風，西南風以及區域風去理解。利用類神經網路所建構出來的多步預測模型的準確率(相關係數 R^2)有達到 0.991，0.981，和 0.970。研究中的參數矯正雖在最後無顯著地提升模型的準確率，但卻可以提高我們對於預測模型的認知。

英文摘要 
In 2016, Taiwan’s government published a new energy policy mandating making Taiwan a nuclearfree island by 2025. Offshore wind energy has become an important project to fill the energy gap after the shutdown of nuclear power plants. The biggest problem with wind power is wind instability. A precise forecasting model not only smooths the operation of wind farms, but also reduce damage to the power grid. In this study, wind data from 2017 to 2019 from the Taipower Meteorological Mast in Chanbin offshore is used in the analysis and model training. A shortterm multistep wind forecasting model is established based on an artificial neural network model and is optimized by changing the model parameters.
The data analysis is divided into three parts: a trend analysis, a statistical analysis, and a spectrum analysis. The trend analysis uses a time series plot and a wind rose plot to describe the trends in wind speed and direction over a threeyear period. The statistical analysis explores the similarities and differences between each year using statistical values. Finally, the spectrogram obtained by the Wavelet Transform explains the features of the different wind sources. Based on the previous analyses, the accuracy and a further error analysis of the forecasting results are discussed. The LSTM model is the main model used in this study. The parameter tuning focuses on the input data in an attempt to obtain better performance from the data rather than from the model.
At the end of this study, we found the wind regime in spring, autumn, and winter to be similar every year and to be easily affected by the southwest monsoon and typhoons in summer. The average wind speed is high in autumn and winter but low in spring and summer. The Wavelet Transform results showed that the wind conditions can be roughly classified into northeast wind, southwest wind, and local wind. The accuracy (R^2 value) of the multistep forecasting model reached 0.991, 0.981, and 0.970, respectively, in the first three step prediction. Although parameter tuning did not significantly improve the accuracy of the forecasting mode, it greatly improved our understanding of the forecasting result.

論文目次 
Table of Contents
Table of Contents i
List of Figures iii
List of Tables vii
1. Introduction 1
1.1. Background and Motivation 1
1.1.1. Taiwan 2
1.1.2. Wind forecasting 5
1.2. Research Objectives and Contributions 6
2. Literature Review 7
2.1. Meteorology 7
2.2. Wind regime in Taiwan 9
2.3. Power Curve 10
2.4. Wind Forecasting 11
3. Materials and Methods 13
3.1. Observation Site and Measurements 13
3.2. Statistical Analysis 15
3.3. Wavelet Spectrum Analysis 18
3.4. Forecasting Methods 20
3.4.1. Persistence Model 20
3.4.2. Artificial Neural Networks 20
3.4.3. Recurrent Neural Network 22
3.5. Neural Network Training Process 24
4. Results and Discussion 27
4.1. Data Sources 27
4.2. Data Analysis 27
4.2.1. Time Series Trends 28
4.2.2. Statistical Analysis 33
4.2.3. Wavelet Spectrum Analysis 37
4.2.4. Section Conclusions 48
4.3. Forecasting Results with the LSTM 49
4.3.1. Data Specifications and Hyperparameter Settings 50
4.3.2. Validation Results 52
4.3.3. Error Analysis 55
4.3.4. Section Conclusion 61
4.4. Forecasting Strategy 62
4.4.1. Model Tests 63
4.4.2. Lookback parameters 66
4.4.3. Lookforward Strategy 67
4.4.4. Wind Direction as a Variable 70
4.4.5. Time Average 73
4.4.6. Different Time Averages to Forecast 30mins 77
4.4.7. Section Conclusion 80
5. Conclusions and Prospects 82
5.1. Conclusions 82
5.2. Prospects 83
6. References 84
7. Appendix 93
7.1. Details of Statistical Properties 93
7.2. Details of Small Scales Wavelet Spectrum in 2017 and 2018 96
7.3. Details of Diurnal Effects 102

參考文獻 
[1] V. Ş. Ediger, "An integrated review and analysis of multienergy transition from fossil fuels to renewables," Energy Procedia, vol. 156, pp. 26, 2019/01/01/ 2019, doi: https://doi.org/10.1016/j.egypro.2018.11.073.
[2] K. Zickfeld, S. Solomon, and D. M. Gilford, "Centuries of thermal sealevel rise due to anthropogenic emissions of shortlived greenhouse gases," Proceedings of the National Academy of Sciences, vol. 114, no. 4, pp. 657662, 2017, doi: 10.1073/pnas.1612066114.
[3] S. Solomon, G.K. Plattner, R. Knutti, and P. Friedlingstein, "Irreversible climate change due to carbon dioxide emissions," Proceedings of the National Academy of Sciences, vol. 106, no. 6, pp. 17041709, 2009, doi: 10.1073/pnas.0812721106.
[4] BP, "BP Statistical Review of World Energy," 2019. [Online]. Available: https://www.bp.com/content/dam/bp/businesssites/en/global/corporate/pdfs/energyeconomics/statisticalreview/bpstatsreview2019fullreport.pdf
[5] BP, "BP Energy Outlook," 2019. [Online]. Available: https://www.bp.com/content/dam/bp/businesssites/en/global/corporate/pdfs/energyeconomics/energyoutlook/bpenergyoutlook2019.pdf
[6] REN21, "Renewables 2019 Global Status Report," 2019. [Online]. Available: https://www.ren21.net/wpcontent/uploads/2019/05/gsr_2019_full_report_en.pdf
[7] I. E. Agency, "Offshore_Wind_Outlook_2019," 2019.
[8] C.C. Wei, P.C. Peng, C.H. Tsai, and C.L. Huang, "Regional Forecasting of Wind Speeds during Typhoon Landfall in Taiwan: A Case Study of WestwardMoving Typhoons," Atmosphere, vol. 9, no. 4, p. 141, 2018. [Online]. Available: https://www.mdpi.com/20734433/9/4/141.
[9] S. Jan, J. Wang, C.S. Chern, and S.Y. Chao, "Seasonal variation of the circulation in the Taiwan Strait," Journal of Marine Systems, vol. 35, no. 3, pp. 249268, 2002/07/01/ 2002, doi: https://doi.org/10.1016/S09247963(02)001306.
[10] 4coffshore, "Global Offshore Wind Speeds Rankings," 2019. [Online]. Available: https://www.4coffshore.com/windfarms/windspeeds.aspx.
[11] E. I. Adminstration. "Analysis  Energy Sector Highlights." Energy Information Adminstration. https://www.eia.gov/international/overview/country/TWN (accessed.
[12] M. o. E. A. Bureau of Energy. "Energy Supply (by Indigenous & Imported)." Bureau of Energy, Ministry of Economic Affairs. https://www.moeaboe.gov.tw/ECW/english/content/ContentLink.aspx?menu_id=1540 (accessed.
[13] Taipower. "電力資訊圖表." Taipower. https://www.taipower.com.tw/tc/Chart.aspx?mid=194 (accessed.
[14] M. o. E. A. Bureau of Energy, "Fouryear Wind Power Promotion Plan," 20191227 2019. [Online]. Available: https://www.moeaboe.gov.tw/ECW/populace/content/ContentDesc.aspx?menu_id=5493.
[15] "Eiger’s Map and Guide to Taiwan’s Offshore Wind Farm Projects." ECS (Asia) Ltd. https://www.eiger.law/windenergy/ (accessed.
[16] MikeEdwards~commonswiki, "Map of the Taiwan Strait," M. o. t. T. Strait, Ed., ed.
[17] E. s. Law, "Eiger’s Map and Guide to Taiwan’s Offshore Wind Farm Projects," E. s. M. a. G. t. T. s. O. W. F. Projects, Ed., ed: Eiger's Law, 2020.
[18] G. N. Bathurst, J. Weatherill, and G. Strbac, "Trading wind generation in short term energy markets," IEEE Transactions on Power Systems, vol. 17, no. 3, pp. 782789, 2002, doi: 10.1109/TPWRS.2002.800950.
[19] J. A. Ging, M. E. Glavin, D. J. Crilly, E. P. Kennedy, P. A. O. Donnell, and J. E. Sustman, "Spatiotemporal analysis of possible wind generation output reductions for the Irish transmission system with a high penetration of renewables," in IET Conference on Renewable Power Generation (RPG 2011), 68 Sept. 2011 2011, pp. 16, doi: 10.1049/cp.2011.0185.
[20] C. Lowery and M. O. Malley, "Impact of Wind Forecast Error Statistics Upon Unit Commitment," IEEE Transactions on Sustainable Energy, vol. 3, no. 4, pp. 760768, 2012, doi: 10.1109/TSTE.2012.2210150.
[21] M. A. Matos and R. J. Bessa, "Setting the Operating Reserve Using Probabilistic Wind Power Forecasts," IEEE Transactions on Power Systems, vol. 26, no. 2, pp. 594603, 2011, doi: 10.1109/TPWRS.2010.2065818.
[22] C. Monteiro et al., "Wind power forecasting : stateoftheart 2009,"; Argonne National Lab. (ANL), Argonne, IL (United States), ANL/DIS101; TRN: US200924%%351 United States 10.2172/968212 TRN: US200924%%351 ANL ENGLISH, 2009. [Online]. Available: https://www.osti.gov/servlets/purl/968212
[23] S. Tewari, C. J. Geyer, and N. Mohan, "A Statistical Model for Wind Power Forecast Error and its Application to the Estimation of Penalties in Liberalized Markets," IEEE Transactions on Power Systems, vol. 26, no. 4, pp. 20312039, 2011, doi: 10.1109/TPWRS.2011.2141159.
[24] S. D. Tomic, "A Study of the Impact of Load Forecasting Errors on Trading and Balancing in a Microgrid," in 2013 IEEE Green Technologies Conference (GreenTech), 45 April 2013 2013, pp. 443450, doi: 10.1109/GreenTech.2013.74.
[25] 薛禹胜 et al., "关于风电不确定性对电力系统影响的评述," 中国电机工程学报, vol. 34, no. 29, pp. 50295040, 2014.
[26] Y. Kawai and A. Wada, "Diurnal sea surface temperature variation and its impact on the atmosphere and ocean: A review," Journal of Oceanography, vol. 63, no. 5, pp. 721744, 2007/10/01 2007, doi: 10.1007/s1087200700630.
[27] Y. Y. Yan, "Land and Sea Breezes," in Encyclopedia of World Climatology, J. E. Oliver Ed. Dordrecht: Springer Netherlands, 2005, pp. 446446.
[28] T. Burton, N. Jenkins, D. Sharpe, and E. Bossanyi, Wind energy handbook. John Wiley & Sons, 2011.
[29] I. Van der Hoven, "Power spectrum of horizontal wind speed in the frequency range from 0.0007 to 900 cycles per hour," Journal of meteorology, vol. 14, no. 2, pp. 160164, 1957.
[30] NikNaks, "A diagram of the atmospheric boundary layer," A. d. o. t. a. b. layer, Ed., ed, 2012.
[31] V. Weather, "Lake  Sea breeze chart. Weather & Climate," L.S. b. c. W. Climate, Ed., ed.
[32] G. R. Energy, "Wind Speed Spectrum," ed.
[33] C. W. Buraeu, "隨季節變換的風—季風." [Online]. Available: https://pweb.cwb.gov.tw/PopularScience/index.php/weather/104%E9%9A%A8%E5%AD%A3%E7%AF%80%E8%AE%8A%E6%8F%9B%E7%9A%84%E9%A2%A8%E2%80%94%E5%AD%A3%E9%A2%A8.
[34] 中央氣象局數位科普網, "風吹流," 季風流, Ed., ed.
[35] C. W. Buraeu, "Flow Regime of Typhoon near Taiwan," F. R. o. T. n. Taiwan, Ed., ed.
[36] V. Sohoni, S. C. Gupta, and R. K. Nema, "A Critical Review on Wind Turbine Power Curve Modelling Techniques and Their Applications in Wind Based Energy Systems," Journal of Energy, vol. 2016, pp. 118, 2016, doi: 10.1155/2016/8519785.
[37] S.C. Chang, "Analysis on Meteorology and Power Generation Characteristics in Chanbin Offshre Area," Department of Mechanical Engineering, National Cheng Kung University, National Cheng Kung University, 2019.
[38] S. S. Soman, H. Zareipour, O. Malik, and P. Mandal, "A review of wind power and wind speed forecasting methods with different time horizons," in North American Power Symposium 2010, 2010: IEEE, pp. 18.
[39] Y. Wu and J. Hong, "A literature review of wind forecasting technology in the world," in 2007 IEEE Lausanne Power Tech, 15 July 2007 2007, pp. 504509, doi: 10.1109/PCT.2007.4538368.
[40] J.S. Hong, "Evaluation of the HighResolution Model Forecasts over the Taiwan Area during GIMEX," Weather and Forecasting, vol. 18, no. 5, pp. 836846, 2003, doi: 10.1175/15200434(2003)018<0836:eothmf>2.0.co;2.
[41] M. Lange and U. Focken, Physical approach to shortterm wind power prediction. Springer.
[42] N. Chen, Z. Qian, I. T. Nabney, and X. Meng, "Wind power forecasts using Gaussian processes and numerical weather prediction," IEEE Transactions on Power Systems, vol. 29, no. 2, pp. 656665, 2013.
[43] Y.L. Hu and L. Chen, "A nonlinear hybrid wind speed forecasting model using LSTM network, hysteretic ELM and Differential Evolution algorithm," Energy Conversion and Management, vol. 173, pp. 123142, 2018.
[44] A. A. Ezzat, M. Jun, and Y. Ding, "SpatioTemporal Asymmetry of Local Wind Fields and Its Impact on ShortTerm Wind Forecasting," IEEE Transactions on Sustainable Energy, vol. 9, no. 3, pp. 14371447, 2018, doi: 10.1109/TSTE.2018.2789685.
[45] E. Erdem and J. Shi, "ARMA based approaches for forecasting the tuple of wind speed and direction," Applied Energy, vol. 88, no. 4, pp. 14051414, 2011, doi: 10.1016/j.apenergy.2010.10.031.
[46] F. Ziel, C. Croonenbroeck, and D. Ambach, "Forecasting wind power–modeling periodic and nonlinear effects under conditional heteroscedasticity," Applied Energy, vol. 177, pp. 285297, 2016.
[47] Z. Niu, Z. Yu, W. Tang, Q. Wu, and M. Reformat, "Wind power forecasting using attentionbased gated recurrent unit network," Energy, vol. 196, p. 117081, 2020.
[48] M. Monfared, H. Rastegar, and H. M. Kojabadi, "A new strategy for wind speed forecasting using artificial intelligent methods," Renewable energy, vol. 34, no. 3, pp. 845848, 2009.
[49] T. Kaur, S. Kumar, and R. Segal, "Application of artificial neural network for short term wind speed forecasting," in 2016 Biennial international conference on power and energy systems: towards sustainable energy (PESTSE), 2016: IEEE, pp. 15.
[50] G. Li and J. Shi, "On comparing three artificial neural networks for wind speed forecasting," Applied Energy, vol. 87, no. 7, pp. 23132320, 2010.
[51] Y. Liu et al., "Wind power shortterm prediction based on LSTM and discrete wavelet transform," Applied Sciences, vol. 9, no. 6, p. 1108, 2019.
[52] H.w. PENG, X.f. YANG, and F.r. LIU, "Shortterm wind speed forecasting of wind farm based on SVM method," Power System and Clean Energy, vol. 7, 2009.
[53] Y. Liu, J. Shi, Y. Yang, and W.J. Lee, "ShortTerm WindPower Prediction Based on Wavelet Transform&Support Vector Machine and StatisticCharacteristics Analysis," vol. 48, no. 4, pp. 11361141, 2012, doi: 10.1109/tia.2012.2199449.
[54] G. SantamaríaBonfil, A. ReyesBallesteros, and C. Gershenson, "Wind speed forecasting for wind farms: A method based on support vector regression," Renewable Energy, vol. 85, pp. 790809, 2016.
[55] W. Zheng et al., "Composite quantile regression extreme learning machine with feature selection for shortterm wind speed forecasting: A new approach," Energy Conversion and Management, vol. 151, pp. 737752, 2017.
[56] X. Luo et al., "Shortterm wind speed forecasting via stacked extreme learning machine with generalized correntropy," IEEE Transactions on Industrial Informatics, vol. 14, no. 11, pp. 49634971, 2018.
[57] M. Abuella and B. Chowdhury, "Random forest ensemble of support vector regression models for solar power forecasting," in 2017 IEEE Power & Energy Society Innovative Smart Grid Technologies Conference (ISGT), 2017: IEEE, pp. 15.
[58] A. Lahouar and J. B. H. Slama, "Hourahead wind power forecast based on random forests," Renewable energy, vol. 109, pp. 529541, 2017.
[59] Y.D. Syu et al., "UltraShortTerm Wind Speed Forecasting for Wind Power Based on Gated Recurrent Unit," in 2020 8th International Electrical Engineering Congress (iEECON), 2020: IEEE, pp. 14.
[60] K. Wang, X. Qi, H. Liu, and J. Song, "Deep belief network based kmeans cluster approach for shortterm wind power forecasting," Energy, vol. 165, pp. 840852, 2018.
[61] Y. Li, H. Shi, F. Han, Z. Duan, and H. Liu, "Smart wind speed forecasting approach using various boosting algorithms, big multistep forecasting strategy," Renewable energy, vol. 135, pp. 540553, 2019.
[62] C. Zhang, H. Wei, J. Zhao, T. Liu, T. Zhu, and K. Zhang, "Shortterm wind speed forecasting using empirical mode decomposition and feature selection," Renewable Energy, vol. 96, pp. 727737, 2016.
[63] K. Chen and J. Yu, "Shortterm wind speed prediction using an unscented Kalman filter based statespace support vector regression approach," Applied Energy, vol. 113, pp. 690705, 2014, doi: 10.1016/j.apenergy.2013.08.025.
[64] S. Tasnim, A. Rahman, A. M. T. Oo, and M. Enamul Haque, "Autoencoder for wind power prediction," Renewables: Wind, Water, and Solar, vol. 4, no. 1, p. 6, 2017/12/16 2017, doi: 10.1186/s408070170044x.
[65] K.P. Lin, P.F. Pai, and Y.J. Ting, "Deep Belief Networks With Genetic Algorithms in Forecasting Wind Speed," IEEE Access, vol. 7, pp. 9924499253, 2019.
[66] H. Zhang, L. Chen, Y. Qu, G. Zhao, and Z. Guo, "Support vector regression based on gridsearch method for shortterm wind power forecasting," Journal of Applied Mathematics, vol. 2014, 2014.
[67] C. Ren, N. An, J. Wang, L. Li, B. Hu, and D. Shang, "Optimal parameters selection for BP neural network based on particle swarm optimization: A case study of wind speed forecasting," Knowledgebased systems, vol. 56, pp. 226239, 2014.
[68] N. Amjady, F. Keynia, and H. Zareipour, "Wind power prediction by a new forecast engine composed of modified hybrid neural network and enhanced particle swarm optimization," IEEE transactions on sustainable energy, vol. 2, no. 3, pp. 265276, 2011.
[69] D. Buttarazzi, G. Pandolfo, and G. C. Porzio, "A boxplot for circular data," Biometrics, vol. 74, no. 4, pp. 14921501, 2018, doi: 10.1111/biom.12889.
[70] K. Gurley and A. Kareem, "Applications of wavelet transforms in earthquake, wind and ocean engineering," Engineering structures, vol. 21, no. 2, pp. 149167, 1999.
[71] L. Cohen, "The Uncertainty Principle for the ShortTime Fourier Transform and Wavelet Transform," in Wavelet Transforms and TimeFrequency Signal Analysis, L. Debnath Ed. Boston, MA: Birkhäuser Boston, 2001, pp. 217232.
[72] M. Stephane, "A wavelet tour of signal processing."
[73] S. G. Mallat, "A theory for multiresolution signal decomposition: the wavelet representation," IEEE transactions on pattern analysis and machine intelligence, vol. 11, no. 7, pp. 674693, 1989.
[74] I. Daubechies, Ten lectures on wavelets. Society for Industrial and Applied Mathematics, 1992.
[75] A. Grossmann and J. Morlet, "Decomposition of Hardy functions into square integrable wavelets of constant shape," SIAM journal on mathematical analysis, vol. 15, no. 4, pp. 723736, 1984.
[76] R. X. Gao and R. Yan, Wavelets: Theory and applications for manufacturing. Springer Science & Business Media, 2010.
[77] R. G. Gragory Lee, Filip Waselewski, Kai Wohlfahrt, Aaron O'Leary, "Pywavelet: A Python package for wavelet analysis," The Journal of Open Source Software, vol. 4, p. 1237, 2019, doi: 10.21105/joss.01237.
[78] A. Teolis and J. J. Benedetto, Computational signal processing with wavelets. Springer, 1998.
[79] W. S. McCulloch and W. Pitts, "A logical calculus of the ideas immanent in nervous activity," The bulletin of mathematical biophysics, vol. 5, no. 4, pp. 115133, 1943.
[80] V. Nair and G. E. Hinton, "Rectified linear units improve restricted boltzmann machines," in Proceedings of the 27th international conference on machine learning (ICML10), 2010, pp. 807814.
[81] L. Lu, Y. Shin, Y. Su, and G. E. Karniadakis, "Dying relu and initialization: Theory and numerical examples," arXiv preprint arXiv:1903.06733, 2019.
[82] D.A. Clevert, T. Unterthiner, and S. Hochreiter, "Fast and accurate deep network learning by exponential linear units (elus)," arXiv preprint arXiv:1511.07289, 2015.
[83] D. P. Kingma and J. Ba, "Adam: A method for stochastic optimization," arXiv preprint arXiv:1412.6980, 2014.
[84] N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, "Dropout: a simple way to prevent neural networks from overfitting," The journal of machine learning research, vol. 15, no. 1, pp. 19291958, 2014.
[85] R. Pascanu, T. Mikolov, and Y. Bengio, "On the difficulty of training recurrent neural networks," in International conference on machine learning, 2013, pp. 13101318.
[86] J. Chung, C. Gulcehre, K. Cho, and Y. Bengio, "Empirical evaluation of gated recurrent neural networks on sequence modeling," arXiv preprint arXiv:1412.3555, 2014.
[87] colah, "Understanding LSTM Networks," S. o. LSTM, Ed., ed, 2015.
[88] C. N. Ciocoiu and R. C. Dobrea, "The role of standardization in improving the effectiveness of integrated risk management," in Advances in Risk Management: IntechOpen, 2010.
[89] Matworks. "goodnessOfFit." https://www.mathworks.com/help/ident/ref/goodnessoffit.html (accessed 2020).
[90] J. Wang, Y. Song, F. Liu, and R. Hou, "Analysis and application of forecasting models in wind power integration: A review of multistepahead wind speed forecasting models," Renewable and Sustainable Energy Reviews, vol. 60, pp. 960981, 2016.
[91] S. B. Taieb, A. Sorjamaa, and G. Bontempi, "Multipleoutput modeling for multistepahead time series forecasting," Neurocomputing, vol. 73, no. 1012, pp. 19501957, 2010.
[92] F. Chollet, "keras," GitHub repository, 2015. [Online]. Available: https://github.com/fchollet/keras.
[93] Wikipedia. "Seasons." https://zh.wikipedia.org/wiki/%E5%AD%A3%E8%8A%82 (accessed.
[94] J. L. Hintze and R. D. Nelson, "Violin plots: a box plotdensity trace synergism," The American Statistician, vol. 52, no. 2, pp. 181184, 1998.
[95] Wikipedia. "2018年強烈颱風鯉魚." https://zh.wikipedia.org/wiki/2018%E5%B9%B4%E9%A2%B1%E9%A2%A8%E6%9D%B0%E6%8B%89%E8%8F%AF

論文全文使用權限 
同意授權校內瀏覽/列印電子全文服務，於20200818起公開。同意授權校外瀏覽/列印電子全文服務，於20210101起公開。 


