臺灣規劃於2025年建置20GW太陽光電發電系統、6.7GW風力發電系統及推動其它再生能源建置以達到再生能源占總發電量20%目標。然而2021年5月臺灣皆於傍晚時間發生兩次大停電，除了人為疏失、氣候變遷及離峰用電日漸遽增因素外，與太陽光電發電系統夜間無法供電及其他發電機組啟動併網需要數小時時間有關，顯見太陽光電發電效益及短期發電量預測與經濟電力調度息息相關。本文主旨在於設計一低成本全自動化太陽能清潔機器人以確保太陽光電發電效益，並研製一智慧型太陽光電發電短期預測機制以精確掌握每小時太陽光電發電量。
由於太陽光電發電原理是藉由陽光照射在太陽光電板進而吸收光能使之發電，當太陽光電板因髒污使得陽光無法完全照射到太陽光電板上時，容易導致發電效率降低。當前唯一能解決該問題的方式，即是擬定相對應的清潔計畫，定時清潔可以有效的去除太陽光電板上的髒汙，使太陽光電發電系統的發電效益維持在一定的水準之上。本文提出一新型太陽能清潔機器人，以低成本且全自動化為目標，不反覆搬運太陽能清潔機器人來降低施工危險性及增加清潔便利性，清潔人員只需要在安全處準備清潔相關設施，透過手機及開關即可遠端對太陽能清潔機器人下達控制命令以達到完整清潔效果。本文所提出太陽能清潔機器人之機構設計簡單，由刷子、側面固定板兩片、三根固定支架、側向輔助輪及馬達組合而成，其操作非常容易，透過實際太陽光電發電系統實測清潔效果並計算發電量改善率，以驗證本文所設計太陽能清潔機器人之有效性，並與市面上產品的相關性能進行比較以顯現本文所提架構之優越性。
為了減少資料傳輸成本，電力營運商的電表資料管理系統通常會延遲時間才取得太陽光電發電系統案場的發電資訊，此作法雖然解決資料傳輸成本的問題，卻對太陽光電發電預測，帶來更高的挑戰性。因電力營運商通常需要即時太陽光電發電量作為電力調度依據，但考量通訊成本無法即時給予對應歷史發電資料，因此本文將提出氣象資料擬合的前處理概念，並結合長短期記憶神經網路模型來進行太陽光電發電量預測，且針對所訓練模型進行實際驗證，將一個完成訓練的模型放置不同太陽光電發電案場來檢測其泛用性。

Taiwan government plans to build 20GW solar photovoltaic (PV) power generation systems, 6.7GW wind power generation systems, and other renewable energy sources in 2025 to achieve the goal of renewable energy accounting for 20% of the total power generation. However, Taiwan experienced two major power outages in the evening in May 2021. In addition to human negligence, climate change, and the increasing off-peak power consumption, the solar PV power generation system cannot provide power at night, and other generators need more times to be reconnected to the utility grid. It is obvious that normal PV power generation and short-term power generation forecasts are closely related to economic power dispatch. The main purpose of this dissertation is to design a low-cost fully-automated solar cleaning robot for ensuring normal solar pv power generation, and to develop an intelligent short-term prediction mechanism of solar PV power generation for accurately obtaining the hourly power generation amount.
The principle of PV power generation is to irradiate the solar panels with sunlight and absorb light energy to generate electricity. When the solar panels are dirty so that the sunlight cannot fully irradiate the solar panels, the power generation efficiency will decrease. At present, the only way to solve this problem is to draw up a corresponding cleaning plan. Regular cleaning can effectively remove the dirt on the PV panel and maintain the power generation efficiency of the PV power generation system above a certain level. This dissertation proposes a new type of solar cleaning robot, aiming at low cost and full automation. It is unnecessary to carry the solar cleaning robot repeatedly to reduce construction risks and increase cleaning convenience. The cleaning staff only needs to prepare related cleaning facilities in a safe place through mobile phones and switches for sending a remote control command to the solar cleaning robot to achieve a complete cleaning effect. The mechanism design of the solar cleaning robot proposed in this dissertation is simple, and its operation is very easy. It is composed of a brush, two side fixed plates, three fixed brackets, lateral auxiliary wheels and motors. The cleaning performance can be verified by actual solar PV power generation systems. The improvement rate of power generation is to verify the effectiveness of the solar cleaning robot designed in this dissertation, and to compare with the relevant performance of products on the market to show the superiority of the architecture proposed in this dissertation.
In order to reduce the cost of data transmission, the meter data management system (MDMS) of the power operator usually delays time to obtain the power generation information of the solar PV power generation system. Although this approach solves the problem of data transmission cost, it brings more challenges to the solar PV power generation forecast. Because power operators usually need real-time solar PV power generation as a basis for the power dispatch, but considering the cost of communication, they cannot always provide corresponding historical power generation data in real time. Therefore, this dissertation will propose a pre-processing concept for weather data fitting combined with a long short-term memory (LSTM) neural network model to predict the amount of solar PV power generation. The effectiveness of the trained model is verified by actual PV power plants, and it places this trained model on different PV power generation cases to test its versatility.

[1] Global energy and CO2 emissions in 2020. Accessed: Apr. 7, 2021. [Online]. Available: https://www.iea.org/reports/global-energy-review-2020/global-energy -and- co2-emissions-in-2020.
[2] World Energy Transitions Outlook: 1.5°C Pathway (Preview). Accessed: Mar. 20, 2021. [Online]. Available: hhttps://www.irena.org/publications/2021/March/ World-Energy-Transitions-Outlook.
[3] 能源轉型白皮書。Accessed: Mar. 25, 2021. [Online]. Available: https:// energywhitepaper.tw/news01.html.
[4] Global PV installations to surpass 150 GW in 2021. Accessed: Mar. 26, 2021. [Online]. Available: https://www.pv-magazine.com/2021/02/16/global-pv- installations -to-surpass-150-gw-in-2021/.
[5] 蘇揆：按部就班 達成2025年太陽光電20GW設置目標。 Accessed: Mar. 26, 2021. [Online]. Available: https://www.ey.gov.tw/Page/9277F759E41CCD91/ db431d06-19f7-4eb5-89e8-fcb0ba1a0869.
[6] F. Creutzig, P. Agoston, J. C. Goldschmidt, G. Luderer, G. Nemet, and R. C. Pietzcker, “The underestimated potential of solar energy to mitigate climate change,” Nat Energy 2, Article 17140, July 2017.
[7] 太陽能板造成環境污染？綠能真相大公開。 Accessed: Mar. 25, 2021. [Online]. Available: https://www.finmart.com.tw/wiki/ALL/solar07.
[8] M. Hammoud, B. Shokr, A. Assi, J. Hallal, and P. Khoury, “Effect of dust cleaning on the enhancement of the power generation of a coastal PV-power plant at Zahrani Lebanon,” Solar Energy, vol. 184, pp. 195-201, Apr. 2019.
[9] A. A. Shehri, B. Parrott, P. Carrasco, H. A. Saiari, and I. Taie, “Impact of dust deposition and brush-based dry cleaning on glass transmittance for PV modules applications,” Solar Energy, vol. 135, pp. 317-324, Oct. 2016.
[10] B. Figgis, A. Nouviaire, Y. Wubulikasimu, W. Javed, B. Guo, A. Ait-Mokhtar, R. Belarbi, S. Ahzi, Y. Rémond, and A. Ennaoui, “Investigation of factors affecting condensation on soiled PV modules,” Solar Energy, vol. 159, pp. 488-500, Jan. 2018.
[11] B. Guo, W. Javed, B. W. Figgis, and T. Mirza, “Effect of dust and weather conditions on photovoltaic performance in Doha, Qatar,” First Workshop on Smart Grid and Renewable Energy (SGRE), Aug. 2015, pp. 1-6.
[12] D. Deb, and N. L. Brahmbhatt, “Review of yield increase of solar panels through soiling prevention, and a proposed water-free automated cleaning solution,” Renewable and Sustainable Energy Reviews, vol. 82, part 3, pp. 3306-3313, Feb. 2018.
[13] S. C.S. Costa, A. S. A.C. Diniz, and L. L. Kazmerski, “Dust and soiling issues and impacts relating to solar energy systems: Literature review update for 2012–2015,” Renewable and Sustainable Energy Reviews, vol. 63, pp. 33-61, Sep. 2016.
[14] M. Saidan, A. G. Albaali, E. Alasis, and J. K. Kaldellis, “Experimental study on the effect of dust deposition on solar photovoltaic panels in desert environment,” Renewable Energy, vol. 92, pp. 499-505, July 2016.
[15] H. Khonkar, A. Alyahya, M. Aljuwaied, M. Halawani, A. A. Saferan, F. Al-khaldi, F. Alhadlaq, and B. A. Wacaser, “Importance of cleaning concentrated photovoltaic arrays in a desert environment,” Solar Energy, vol. 110, pp. 268-275, Dec. 2014.
[16] A. A. Shehri, B. Parrott, P. Carrasco, H. A. Saiari, and I. Taie, “Impact of dust deposition and brush-based dry cleaning on glass transmittance for PV modules applications,” Solar Energy, vol. 135, pp. 317-324, Oct. 2016.
[17] A. Kimber, L. Mitchell, S. Nogradi, and H. Wenger, “The effect of soiling on large grid-connected photovoltaic systems in California and the southwest region of the United States,” IEEE 4th World Conference on Photovoltaic Energy Conference, Waikoloa, HI, USA, May 2006, pp. 2391-2395.
[18] M. G. Deceglie, L. Micheli, and M. Muller, “Quantifying soiling loss directly from PV yield,” IEEE Journal of Photovoltaics, vol. 8, no. 2, pp. 547-551, Mar. 2018.
[19] M. G. Deceglie, M. Muller, Z. Defreitas, and S. Kurtz, “A scalable method for extracting soiling rates from PV production data,” IEEE 43rd Photovoltaic Specialists Conference (PVSC), Portland, OR, USA, June 2016, pp. 2061-2065.
[20] Å. Skomedal, H. Haug, and E. S. Marstein, “Endogenous soiling rate determination and detection of cleaning events in utility-scale PV plants,” IEEE Journal of Photovoltaics, vol. 9, no. 3, pp. 858-863, May 2019.
[21] M. Gostein, B. Littmann, J. R. Caron, and L. Dunn, “Comparing PV power plant soiling measurements extracted from PV module irradiance and power measurements,” IEEE 39th Photovoltaic Specialists Conference (PVSC), Tampa, FL, USA, June 2013, pp. 3004-3009.
[22] F. A. Mejia, and J. Kleissl, “Soiling losses for solar photovoltaic systems in California,” Solar Energy, vol. 95, pp. 357-363, Sep. 2013.
[23] B. Parrott, P. C. Zanini, A. Shehri, K. Kotsovos, and I. Gereige, “Automated, robotic dry-cleaning of solar panels in Thuwal, Saudi Arabia using a silicone rubber brush,” Solar Energy, vol. 171, pp. 526-573, Sep. 2018.
[24] 智能運維機器人。Accessed: Mar. 27, 2021. [Online]. Available: http:// www.solbrighttech.com/index.php?c=article&a=type&tid=9.
[25] 第三代可攜式太陽能板清洗機ST001。Accessed: Mar. 27, 2021. [Online]. Available: https://www.facebook.com/DongYicore/
[26] 太陽能模組清洗設備。 Accessed: Mar. 27, 2021. [Online]. Available: https:// www.de-tools.com.tw/collections/isolar.
[27] 智慧清潔遙控機器人。 Accessed: Mar. 27, 2021. [Online]. Available: http:// www.worldhigh.com.tw/modulecleaning.html.
[28] WET PATROLLER. Accessed: Mar. 27, 2021. [Online]. Available: https:// www.geva-bot.com/robots-1/%D7%93%D7%92%D7%9D-1000.
[29] 再生能源是台灣永續發展的王道。Accessed: Apr. 20, 2021. [Online]. Available: https://view.ctee.com.tw/environmental/12560.html.
[30] E. Cirés, J. Marcos, I. de la Parra, M. García, and L. Marroyo, “The potential of forecasting in reducing the LCOE in PV plants under ramp-rate restrictions,” Energy, vol. 188, Article 116053, Dec. 2019.
[31] Solar set for 209GW surge in 2021. Accessed: Mar. 27, 2021. [Online]. Available: https://renews.biz/66679/solar-capacity-installations-in-2021-could-hit-209gw/.
[32] 綠能科技產業創新推動方案。Accessed: Apr. 15, 2021. [Online]. Available: https://www.ey.gov.tw/Page/448DE008087A1971/8484fe7c-6e0c-4f4b-a79f-4e9b6076c2bc.
[33] 再生能源統計資料-再生能源發電量統計110年1月份月報。Accessed: Apr. 15, 2021. [Online]. Available: https://www.re.org.tw/information/statistics_more. aspx?id=3879.
[34] 再生能源發電概況。Accessed: Apr. 10, 2021. [Online]. Available: https://www.taipower.com.tw/tc/page.aspx?mid=204&cid=1582&cchk=5b8ce619-7ff5-40e9-9032-bdfd93d197d9.
[35] 如何挑選太陽能模組技術？Accessed: Apr. 13, 2021. [Online]. Available: https://www.winaico.com/tw/blog/solar-panel-types/.
[36] 晶矽組件和雙結矽基薄膜組件的優缺點分析。Accessed: Apr. 13, 2021. [Online]. Available: https://kknews.cc/zh-tw/finance/9pgayq.html.
[37] 【再生能源知多少】 太陽能發電。Accessed: Apr. 15, 2021. [Online]. Available:https://www.youtube.com/watch?v=PcKzeUrgwxs&ab_channel=%E5%8F%B0%E9%9B%BB%E5%BD%B1%E9%9F%B3%E7%B6%B2.
[38] 太陽能光電系統主要類型。Accessed: Apr. 15, 2021. [Online]. Available: https://www.pvesco168.com.tw/solartype.html.
[39] 太陽能系統裝置。Accessed: Apr. 15, 2021. [Online]. Available:http://www.opti-solar.com/taiwan/.
[40] 太陽能成本回收需要多少時間？參與太陽能理財的成本與風險分析。Accessed: Apr. 20, 2021. [Online]. Available:https://www.finmart.com.tw/ Wiki/ALL/article45.
[41] L. Golubchik, S. Khuller, K. Mukherjee, and Y. Yao, “To send or not to send: Reducing the cost of data transmission,” Proc. IEEE INFOCOM, Apr. 2013, pp. 2472-2478.
[42] 台電資安防護及再生能源發電預測系統，預算4,725萬。Accessed: Apr. 12, 2021. [Online]. Available: https://www.govtechnews.org/post/台電資安防護及再生能源發電預測系統，預算4-725萬.
[43] 大量再生能源併網的衝擊與電力調度因應策略。Accessed: Apr. 12, 2021. [Online]. Available: http://www.cie.org.tw/cms/JournalFiles/10903_chapter09. pdf.
[44] K. Porter, S. Fink, M. Buckley, J. Rogers, and B. M. Hodge, “Review of variable generation integration charges,” Nat. Renewable Energy Lab.(NREL), Golden, CO, USA, Tech. Rep. NREL/TP-5500-57583, Mar. 2013.
[45] D. Lew, N. Piwko, D. Miller, G. Jordan, K. Clark, and L. Freeman, “NREL: How do high levels of wind and solar impact the grid? The western wind and solar integration study,” Nat. Renewable Energy Lab.(NREL), Golden, CO, USA, Tech. Rep. NREL/TP-5500-50057, Dec. 2010.
[46] 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.
[47] 淺談新能源市場現況(中)：再生能源供給的挑戰。Accessed: Apr. 15, 2021. [Online]. Available: https://www.digitimes.com.tw/col/article.asp?id=1125.
[48] Negative electricity prices: lockdown’s demand slump exposes inflexibility of German power. Accessed: Apr. 18, 2021. [Online]. Available: https://energypost.eu/negative-electricity-prices-lockdowns-demand-slump-exposes-inflexibility-of-german-power/.
[49] Coronavirus slump will lead to more hours of negative power prices in Germany –study. Accessed: Apr. 18, 2021. [Online]. Available: https://www.cleanenergywire.org/news/coronavirus-slump-will-lead-more-hours-negative-power-prices-germany-study.
[50] Germany's negative price rules bring negative consequences. Accessed: Apr. 13, 2021. [Online]. Available: https://www.districtenergy.org/blogs/ district-energy/2020/05/01/germanys-negative-price-rules-bring-negative-conse.
[51] J. C.Ketterer, “The impact of wind power generation on the electricity price in Germany,” Energy Economics, vol. 44, pp. 270-280, July 2014.
[52] K. D. Vos, “Negative wholesale electricity prices in the German, French and Belgian day-ahead, intra-day and real-time markets,” The Electricity Journal, vol. 28, pp. 36-50, May 2015.
[53] 德國負電價趨勢對風場收益有何影響? 淺談再生能源發電補貼抑制條款。Accessed: Apr. 16, 2021. [Online]. Available: https://www.infolink-group.com/ zh-tw/wind/wind-analysis/how-does-the-negative-electricity-price-affect-wind-farm-revenue-in-germany.
[54] 中國棄風棄光 對台灣再生能源有什麼啟示？Accessed: Apr. 16, 2021. [Online]. Available: https://www.cw.com.tw/article/5079045.
[55] China’s 2019 electricity generation reviewed as its next 5-year plan is drafted. Accessed: Apr. 16, 2021. [Online]. Available: https://energypost.eu/ chinas-2019-electricity-generation-reviewed-as-its-next-5-year-plan-is-drafted/.
[56] Green power curtailment in China. Accessed: Apr. 12, 2021. [Online]. Available: https://physicsworld.com/a/green-power-curtailment-in-china/.
[57] N. Tang, Y. Zhang, Y. Niu, and X. Du, “Solar energy curtailment in China: Status quo, reasons and solutions,” Renewable and Sustainable Energy Reviews, vol. 97, pp. 509-528, Dec. 2018.
[58] C. Li, H. Shi, Y. Cao, J. Wang, Y. Kuang, Y. Tan, and J. Wei, “Comprehensive review of renewable energy curtailment and avoidance: A specific example in China,” Renewable and Sustainable Energy Reviews, vol. 41, pp. 1067-1079, Jan. 2015.
[59] D. Yang and Z. Dong, “Operational photovoltaics power forecasting using seasonal time series ensemble,” Solar Energy, vol. 166, pp. 529-541, May 2018.
[60] G. Wang, Y. Su, and L. Shu, “One-day-ahead daily power forecasting of photovoltaic systems based on partial functional linear regression models,” Renewable Energy, vol. 96, pp. 469-478, Oct. 2016.
[61] M. Kang, J. Sohn, J. Park, S. Lee, and Y. Yoon, “Development of algorithm for day ahead PV generation forecasting using data mining method,” IEEE 54th Int. Midwest Symposium on Circuits and Systems (MWSCAS), Aug. 2011, pp. 1-4.
[62] Y. Kashyap, A. Bansal, and A. K.Sao, “Solar radiation forecasting with multiple parameters neural networks,” Renewable and Sustainable Energy Reviews, vol. 49, pp. 825-835, Sept. 2015.
[63] V. Kushwaha and N. M. Pindoriya, “Very short-term solar PV generation forecast using SARIMA model: A case study,” 7th Int. Conf. Power Systems (ICPS), Pune, India, Dec. 2017, pp. 430-435.
[64] M. Pierro, F.Bucci, M. D. Felice, E. Maggioni, D. Moser, A. Perotto, F. Spada, and C. Cornaro, “Multi-model ensemble for day ahead prediction of photovoltaic power generation,” Solar Energy, vol. 134, pp. 132-146, Sept. 2016.
[65] C. Yang, A. A. Thatte, and L. Xie, “Multitime-scale data-driven spatio-temporal forecast of photovoltaic generation,” IEEE Trans. Sustainable Energy, vol. 6, no. 1, pp. 104-112, Jan. 2015.
[66] X. Chen, Y. Du, H. Wen, L. Jiang, and W. Xiao, “Forecasting-based power ramp-rate control strategies for utility-scale PV systems,” IEEE Trans. Ind. Electron., vol. 66, no. 3, pp. 1862-1871, Mar. 2019.
[67] L. Massidda and M. Marrocu, “Use of multilinear adaptive regression splines and numerical weather prediction to forecast the power output of a PV plant in Borkum, Germany,” Solar Energy, vol. 146, pp. 141-149, Apr. 2017.
[68] G. Reikard, S. E. Haupt, and T. Jensen, “Forecasting ground-level irradiance over short horizons: Time series, meteorological, and time-varying parameter models,” Renewable Energy, vol. 112, pp. 474-485, Nov. 2017.
[69] A. Mellit, A. M. Pavan, and V. Lughi, “Short-term forecasting of power production in a large-scale photovoltaic plant,” Solar Energy, vol. 105, pp. 401-413, July 2014.
[70] 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,” Mathematics and Computers in Simulation, vol. 131, pp. 88-100, Jan. 2017.
[71] S. P. Durrani, S. Balluff, L. Wurzer, and S. Krauter, “Photovoltaic yield prediction using an irradiance forecast model based on multiple neural networks,” J. Modern Power Systems and Clean Energy, vol. 6, no. 2, pp. 255-267, Mar. 2018.
[72] A. Alfadda, S. Rahman, and M. Pipattanasomporn, “Solar irradiance forecast using aerosols measurements: A data driven approach,” Solar Energy, vol. 170, pp. 924-939, Aug. 2018.
[73] Y. Sun, V. Venugopal, and A. R. Brandt, “Convolutional neural network for short-term solar panel output prediction,” IEEE 7th World Conf. Photovoltaic Energy Conversion (WCPEC) (A Joint Conference of 45th IEEE PVSC, 28th PVSEC & 34th EU PVSEC), Waikoloa, HI, USA, June 2018, pp. 2357-2361.
[74] Y. Yu, J. Cao, X. Wan, F. Zeng, J. Xin, and Q. Ji, “Comparison of short-term solar irradiance forecasting methods when weather conditions are complicated,” J. Renew. Sustain. Energy, vol. 10, no. 5, Sep. 2018.
[75] A. Rahman, V. Srikumar, and A. D. Smith, “Predicting electricity consumption for commercial and residential buildings using deep recurrent neural networks,” Applied Energy, vol. 212, pp. 372-385, Feb. 2018.
[76] P. J. Tang, H. L. Wang, and K. S. Xu, “Multi-objective layer-wise optimization and multi-level probability fusion for image description generation using LSTM,” Acta Automatica Sinica, vol. 40, no. 7, pp. 1237-1249, July 2018.
[77] L. Yang, Y. Zheng, X. Cai, H. Dai, D. Mu, L. Guo, and T. Dai, “A LSTM based model for personalized context-aware citation recommendation,” IEEE Access, vol. 6, pp. 59618-59627, Oct. 2018.
[78] H. Zhou, Y. Zhang, L. Yang, Q. Liu, K. Yan, and Y. Du, “Short-term photovoltaic power forecasting based on long short term memory neural network and attention mechanism,” IEEE Access, vol. 7, pp. 78063-78074, June 2019.
[79] Y. Yu, J. Cao, and J. Zhu, “An LSTM short-term solar irradiance forecasting under complicated weather conditions,” IEEE Access, vol. 7, pp. 145651-145666, Oct. 2019.
[80] T. Yang, B. Li, and Q. Xun, “LSTM-attention-embedding model-based day-ahead prediction of photovoltaic power output using Bayesian optimization,” IEEE Access, vol. 7, pp. 171471-171484, Nov. 2019.
[81] M. S. Hossain and H. Mahmood, “Short-term photovoltaic power forecasting using an LSTM neural network and synthetic weather forecast,” IEEE Access, vol. 8, pp. 172524-172533, Sept. 2020.
[82] Y. Zhang, C. Qin, A. K. Srivastava, C. Jin, and R. K. Sharma, “Data-driven day-ahead PV estimation using autoencoder-LSTM and persistence model,” IEEE Trans. Ind. Applicat., vol. 56, no. 6, pp. 7185-7192, Sept. 2020.
[83] C. H. Liu, J. C. Gu, and M.T. Yang, “A simplified LSTM neural network for one day-ahead solar power forecasting,” IEEE Access, vol. 9, pp. 17174-17195, Jan. 2021.
[84] 如何正確認識光伏組件中的「熱斑效應」。 Accessed: Mar. 25, 2021. [Online]. Available: https://www.google.com.tw/amp/s/kknews.cc/news/veg2eql.amp.
[85] 光伏板面衰减發電量少的原因。 Accessed: Mar. 25, 2021. [Online]. Available: https://www.kstar.com.cn/articles/16081.jhtml.
[86] 你真的知道嗎？遮陰對模組的影響。Accessed: Mar. 25, 2021. [Online]. Available: https://www.initialsolar.com/en/blog/221.
[87] 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.
[88] 如何正確認識光伏組件中的「熱斑效應」。 Accessed: Mar. 25, 2021. [Online]. Available: https://www.google.com.tw/amp/s/kknews.cc/news/veg2eql.amp.
[89] H. Zhan, Q. Li, K. Zhao, L. Zhang, Z. Zhang, C. Zhang, and L. Xiao, “Evaluating PM2.5 at a construction site using terahertz radiation,” IEEE Trans. Terahertz Science and Technology, vol. 5, no. 6, pp. 1028-1034, Nov. 2015.
[90] C. Zhang, W. Li, J. Yu, and R. Xu, “Modeling impacts of PM 2.5 concentration on PV power outputs,” Int. Conf. Probabilistic Methods Applied to Power Systems (PMAPS), Beijing, China, Oct. 2016, pp. 1-7.
[91] I. M. Peters, S. Karthik, T. Buonassisi, and A. Nobre, “The impact of fine particular matter (PM2.5) on photovoltaic energy yield - A study on the example of New Delhi,” IEEE 7th World Conf. Photovoltaic Energy Conversion (WCPEC), Waikoloa, HI, USA, June 2018, pp. 2581-2583.
[92] 懸浮微粒。Accessed: Mar. 28, 2021. [Online]. Available: https://zh. wikipedia.org/wiki/%E6%87%B8%E6%B5%AE%E7%B2%92%E5%AD%90.
[93] 關於細懸浮微粒 PM2.5。Accessed: Mar. 28, 2021. [Online]. Available: https:// www.mirdc.org.tw/FileDownLoad/HealthNews/201721410614879.pdf.
[94] 2020全球空氣質量報告。Accessed: Mar. 28, 2021. [Online]. Available: https:// www.iqair.com/tw/world-most-polluted-countries.
[95] 太陽能板清洗方式。Accessed: Mar. 25, 2021. [Online]. Available: https:// www.winaico.com/tw/guides/cleaning-solar-panel/.
[96] 太陽能板的清潔好幫手【E-PPA】。Accessed: Mar. 25, 2021. [Online]. Available: https://www.chen-yang.cc/news/detail/24 .
[97] UCFL-連座軸承。Accessed: May. 12, 2021. [Online]. Available: https://www.cyrus-linear.com/product-detail.php?id=189&lang=tw
[98] PWM 10A dc motor controller forward and backward linear actuator governor speed control self-reset 6V/12V/24V. Accessed: May 13, 2021. [Online]. Available: https://www.aliexpress.com/item/4000111119961.html
[99] DC 6-24V 3A PWM DC Motor Speed Controller Forward Backward Reversible Switch RC Control Regulator With Relay. Accessed: May 13, 2021. [Online]. Available: https://www.aliexpress.com/item/4000340456196.html
[100] C. Cortes and V. Vapnik, “Support-vector networks,” Machine Learning, vol. 20, pp. 273-297, Sept. 1995.
[100] S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neural Computation, vol. 9, pp. 1735-1780, Dec. 1997.
[101] A. Alzahrani, P. Shamsi, C. Dagli, and M. Ferdowsi, “Solar irradiance forecasting using deep neural networks,” Procedia Computer Science, vol. 114, pp. 304-313, Sept. 2017.
[102] D. E. Rumelhart, G. E. Hinton and R. J. Williams, “Learning representations by back-propagating errors,” Nature 323, vol. 9, pp. 533-536, Oct. 1986.
[103] E. Ostertagová, “Modelling using polynomial regression,” Procedia Engineering, vol. 48, pp. 500-506, Sept. 2012.
[104] G. B. Huang, Q. Y. Zhu, and C. K. Siew, “Extreme learning machine: Theory and applications,” Neurocomputing, vol. 70, pp. 489-501, Dec. 2006.
[105] 台電工程月刊 856 期（12 月號）：再生能源之互通性研發與應用－民營再生能源發電即時資訊平台示範應用案例。Accessed: July 12, 2021. [Online]. Available: https://www.taipower.com.tw/tc/report.aspx?pn=2&mid=86.