世界各國持續推動再生能源發展，太陽光電發電系統的設置量也持續上升。台灣由於天然資源的缺乏，絕大多數能源來源都是仰賴進口，但價格易受到國際市場的波動影響，並有被控制的風險。而台灣電力市場隨著經濟發展，對電力的需求也持續上升，為了要增加整體發電量與分散風險，政府目前也在積極推廣設置再生能源。除了能夠增加發電量外，對環境也更加友善，太陽光電又為其重要的發展項目。目前台灣教育體系配合政府政策也持續推動太陽光電發電系統的新設，除鼓勵學校機關自建外，同步推行太陽光電能源技術服務業，降低學校機關資金負擔與設備維護管理的問題。本文整理目前台灣教育體系推動太陽光電發電發展的近況，並描述推動方法與過程中會遇到的問題，說明學校機關辦理自行標租與參與聯合標租各自的優劣並說明其推動結果。
隨著太陽光電發電系統在電力網絡中的滲透率持續上升，太陽光電發電系統所產生電力的波動與間歇性問題對電力系統的影響將會日益嚴重，如果能更有效預測太陽光電發電系統發電量可降低對電網的影響。以往研究太陽光電發電量預測往往都需要使用外部的天氣預報或是於太陽光電發電場域增設額外的環境感測器才能提供準確的預測。本文發展一利用太陽光電發電場域使用的逆變器內部參數即可獲得準確的太陽能發電量預測模型。資料輸入部分利用相關性分析找出環境因素與逆變器內部參數之間的關聯性，簡化輸入參數的來源與降低取得難易度，再利用具滑動視窗的長短期記憶神經網路模型進行每小時發電量預測。本文所提出的預測模型是採用線上學習，所以在單一場域訓練完畢後也能將模型轉移至其它場域，經數值模擬測試後證明具滑動視窗的長短期記憶神經網路模型能夠於不同場域準確預估太陽光電發電量，並探討此預測模型未來與其它應用結合發展可行性。

Renewable energy is the fastest-growing energy source in the countries around the world, and the worldwide installed capacity of solar photovoltaic (PV) systems is growing rapidly. Most of the energy resources used in Taiwan is import from oversea because of the nature resource scarcity. The price of energy sources is controlled by international situation, and the cost is expensive and unexpected. Due to economic development in Taiwan, the power demand is rising rapidly. In order to increase the power generation and diversify risks, the government of Taiwan promotes the installation of renewable energy in recent years. Renewable energy not only increases power generations, but also is more environment friendly. The solar PV system is one of the most important energy in renewable energies. At present, Taiwan education systems follow the government policies for promoting the installation of PV systems. In order to encourage schools or institutions to have their own PV plants, the cooperation with PV energy service company (PV-ESCO) is the other way. The PV-ESCO can reduce the financial burden, and deals with equipment maintenance and management issues to schools or institutions. In this thesis, it summarizes current status of promoting the installation of solar PV systems in Taiwan education system. Moreover, it describes the promoting method and the most common problems, and explains the different between own bidding and joint bidding.
As the penetration rate of PV systems in the power grid is rising, the fluctuations and intermittent problems of the electricity generated from PV systems will affect the power system quality. If there is a more effectively prediction for the PV power generation, it can reduce the impact to the power grid. In general, the forecasting of PV power generation in previous researches always requires external weather forecasts or installs additional environmental sensors in the PV plant to provide the more accurate forecasting of PV power generation. This thesis proposes an accurate prediction model for the PV power generation by using internal parameters from PV inverters, which can be easily obtained from each PV plant. The data input uses correlation analyses to find the correlation between environmental factors and internal parameters of the inverter to simplifying the input data amount. In addition, it combines a long-short-term-memory (LSTM) neural network model and a sliding windows method to predict the hourly PV power generation. The model proposed in this thesis is also an online learning model, such that the model can be trained via the data from one PV plant, and is easily to use in different PV plants. Numerical simulations via real data from different PV systems are verifed the effectiveness of the proposed sliding-windows LSTM (SW-LSTM) model. Futhermore, it explores the feasibility of combining this prediction model with other applications in the future.

[1] E. C. Marcos, I. Parra, M. Garcia, and L. Marroyo, “The potential of forecasting in reducing the LCOE in PV plants under ramp-rate restrictions,” Energy, vol. 188, no.1, Article 116053, Dec. 2019.
[2] P. Li, K. Zhou, X. Lu, and S. Yang, “A hybrid deep learning model for short-term PV power forecasting,” Applied Energy, https://doi.org/10.1016/j.apenergy.2019.114216.
[3] Y. Han, N. Wang, M . Ma, H. Zhou, S. Dai, and H. Zhu ” A PV power interval forecasting based on seasonal model and nonparametric estimation algorithm,” Solar Energy, vol. 184, pp. 515-526, May 2019.
[4] M. Marinelli, F. Sossan, G. T. Costanzo, and H. W. Bindner, “Testing of a predictive control strategy for balancing renewable sources in a microgrid,” IEEE Trans. Sustain. Energy, vol. 5, no. 4, pp. 1426-1433, Oct. 2014.
[5] M. Tucci, E. Crisostomi, G. Giunta, and M. Raugi, “A multi-objective method for short-term load forecasting in European countries,” IEEE Trans. Power Syst., vol. 31, no. 5, pp. 3537-3547, Sept. 2016.
[6] A. Yona, T. Senjyu, T. Funabashi, and C. H. Kim, “Determination methodof insolation prediction with fuzzy and applying neural network for longterm ahead PV power output correction,” IEEE Trans. Sustain. Energy, vol. 4, no. 2, pp. 527-533, April 2013.
[7] J. M. Huang, R. J. Wai, and W. Gao, “Newly-designed fault diagnostic method for solar photovoltaic generation system based on IV-curve measurement,” IEEE Access, vol.7, pp. 70919-70932, May 2019.
[8] H. Shaker, D. Manfre and H. Zareipour, ” Forecasting the aggregated output of a large fleet of small behind-the-meter solar photovoltaic sites,” Renewable Energy, vol. 147, pp. 1861-1869, Mar. 2020.
[9] D. Yang and Z. Dong, ” Operational photovoltaics power forecasting using seasonal time series ensemble,” Solar Energy, vol. 166, pp. 529-541, May 2018.
[10] P. Bacher, H. Madsen and H. A. Nielsen, “Online short-term solar power forecasting”, Sol. Energy, vol. 83, no. 10, pp. 1772-1783, 2009.
[11] C. Yang, A. Thatte, and L. Xie, “Multitime-scale data-driven spatio-temporal forecast of photovoltaic generation,” IEEE Trans. Sustain. Energy, vol. 6, no. 1, pp. 104-112, Jan. 2015.
[12] J. G. S. Fonseca Jr., T. Oozeki, H. Ohtake, K. Shimose, T. Takashima, and K. Ogimoto, “Regional forecasts and smoothing effect of photovoltaic power generation in Japan: An approach with principal component analysis,” Renewable Energy, vol. 68, pp. 403-413, Aug. 2014.
[13] R. J Bessa, A. Trindade and V. Miranda, “Spatial-temporal solar power forecasting for smart grids,” IEEE Trans. Ind. Inf., vol. 11, pp. 232-241, Feb. 2015.
[14] M. Bouzerdoum, A. Mellit, and A. M. Pavan, “A hybrid model (SARIMA–SVM) for short-term power forecasting of a small-scale grid-connected photovoltaic plant,” Sol. Energy, vol. 98, pp. 226-235, Dec. 2013.
[15] D. P. Larson, L. Nonnenmacher, and C. F. M. Coimbra, “Day-ahead forecasting of solar power output from photovoltaic plants in the American Southwest,” Renewable Energy, vol. 91, pp. 11-20, 2016.
[16] Y. Li, Y. Su, and L. Shu, ”An ARMAX model for forecasting the power output of a grid connected photovoltaic system,” Renewable Energy, vol. 66, pp. 78-79, Jun. 2014.
[17] M. P. Almeida, O. Perpinan, and L. Narvarte, “PV power forecast using a nonparametric PV model,” Solar Energy, vol. 115, pp. 354-368, May 2015.
[18] M. Malvoni, M. G. Giorgi, and P. M. Congedo, “Data on support vector machines (SVM) model to forecast photovoltaic power,” Data in Brief, vol. 9, pp. 13-16, Dec. 2016.
[19] S. Leva, A. Dolara, F. Grimaccia, M. Mussetta, and E. Ogliari, “Analysis and validation of 24 hours ahead neural network forecasting of photovoltaic output power,” Math. Computers in Simulation, vol. 131, pp. 88-100, Jan. 2017.
[20] M. Abuella, and B. Chowdhury, “Solar power forecasting using artificial neural networks,” Proc. North Amer. Power Symp. (NAPS), pp. 1-5, Oct. 2015.
[21] A. Chaouachi, R. M. Kamel, R. Ichikawa, H. Hayashi, and K. Nagasaka, “Neural network ensemble based solar power generation short-term forecasting,” Int. J. Inf. Math. Sci., pp. 332-337, 2009.
[22] R. Muhammad Ehsan, S. P. Simon, and P. R. Venkateswaran, “Dayahead forecasting of solar photovoltaic output power using multilayer perceptron,” Neural Comput. Appl., vol. 28, no. 12, pp 3981-3992, Dec. 2017.
[23] Z. Li, C. Zang, P. Zeng, H. Yu, and H. Li, “Day-ahead hourly photovoltaic generation forecasting using extreme learning machine,” Proc. Annu. IEEE Int. Conf. Cyber Technol. Autom., Control Intell. Syst., pp. 779-783, 2015.
[24] F. Golestaneh, P. Pinson, and H. B. Gooi, “Very short-term nonparametric probabilistic forecasting of renewable energy generation with application to solar energy,” IEEE Trans. Power Syst., vol. 31, pp. 3850-3863, Sept. 2016.
[25] G. Cervone, L. Clemente-Harding, S. Alessandrini, and L. D. Moncache, “Short-term photovoltaic power forecasting using artificial neural networks and an analog ensemble,” Renewable Energy, vol. 108, pp. 274-286, Aug. 2017.
[26] C. D. Dumitru, A. Gligor, and C. Enachescu, “Solar photovoltaic energy production forecast using neural networks,” Procedia Technol., vol. 22, pp. 808-815, Jan. 2016.
[27] M. Fidan, F. O. Hocaoğlu, and Ö. N. Gerek, “Harmonic analysis based hourly solar radiation forecasting model,” IET Renewable Power Generation, vol. 9, pp. 218-228, April 2015.
[28] D. AlHakeem, P. Mandal, A. U. Haque, A. Yona, T. Senjyu, and T. Tseng, “A new strategy to quantify uncertainties of wavelet-GRNN-PSO based solar PV power forecasts using bootstrap confidence intervals,” 2015 IEEE Power & Energy Society General Meeting, Denver, CO, USA, July 2015.
[29] M. Abdel-Nasser and K. Mahmoud, “Accurate photovoltaic power forecasting models using deep LSTM-RNN,” Neural Comput. Appl., pp. 1-14, Oct. 2017.
[30] 能源供給(原始單位)(月)，經濟部能源局，能源統計月報(2019年 12月13日)，https://www.moeaboe.gov.tw/ECW/populace/web_book /WebReports.aspx?book=M_CH&menu_id=142。
[31] 能源轉型白皮書，經濟部，https://energywhitepaper.tw/。
[32] 區域性儲能設備技術示範驗證計畫，經濟部能源局，https://www. ey.gov.tw/File/953538E78B9A4F34。
[33] W. Choi, W. Lee, D. Han and B. Sarlioglu, ” New Configuration of Multifunctional Grid-Connected Inverter to Improve Both Current-Based and Voltage-Based Power Quality,” IEEE Trans. Ind. Appl., vol.54, pp. 6374-6382, July 2018.
[34] 再生能源發電概況，台灣電力公司，https://www.taipower.com.tw/ tc/page.aspx?mid=204 。
[35] G. Cybenko, “Neural networks in computational science and engineering,” IEEE Comput. Sci. and Eng., vol. 3, pp.36-42, Spr. 1996.
[36] C. Cortes and V. Vapnik, ”Support-vector networks,” Machine Learning, vol. 20, pp. 273-297, 1995.
[37] C. Yan, C. Zhang, and Y. Ma, “Study on regression analysis and prediction of time-series data streams using sliding windows,” J. Natural Sci. of Heilongjiang Univ., vol. 23, no. 6, pp. 863-867, 2007.
[38] S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neural Comput., pp. 1735-80, Dec. 1997.
[39] T. Meyer and J. Luther, “On the correlation of electricity spot market prices and photovoltaic electricity generation,” Energy Conversion and Mgmt., vol. 45, pp. 2639-2644, Oct 2004.
[40] J. Ospina, A. Newaz, and M. O. Faruque, “Forecasting of PV plant output using hybrid wavelet-based LSTM-DNN structure model,” IET Renewable Power Generation, vol. 13, pp. 1087-1095, May 2019.
[41] H. Zang, L. Cheng, T. Ding, K. W. Cheung, Z. Liang, Z. Wei, and G. Sun, “Hybrid method for short-term photovoltaic power forecasting based on deep convolutional neural network,” IET Generation, Transmission & Distribution, vol. 12, pp. 4557-4567, Nov. 2018.
[42] M. Yang and X. Huang, “Ultra-short-term prediction of photovoltaic power based on periodic extraction of PV energy and LSH algorithm,” IEEE Access, vol. 6, pp. 51200-51205, Sept. 2018.
[43] A. Klingler and L. Teicgtmann, “Impacts of a forecast-based operation strategy for grid-connected PV storage systems on profitability and the energy system,” Solar Energy, vol. 158, pp. 861-868, Dec. 2017.
[44] L.Wen, K. Zhou, S. Yang and X. Lu, “Optimal load dispatch of community microgrid with deep learning based solar power and load forecasting,” Energy, vol.171, pp. 1053-1065, Mar. 2019.