簡易檢索 / 詳目顯示

回結果列表
研究生： 許哲源
Hsu, Che-Yuan
論文名稱： 基於LSTM系集自我學習方法之極短期太陽能發電預測
LSTM Based Ensemble Self-learning Approach for Very Short-Term PV Generation Forecasting
指導教授： 楊宏澤
Yang, Hong-Tzer
學位類別： 碩士
Master
系所名稱： 電機資訊學院 - 電機工程學系
Department of Electrical Engineering
論文出版年： 2021
畢業學年度： 109
語文別： 英文
論文頁數： 53
中文關鍵詞： 短期太陽能發電預測 、能源管理系統 、長短期記憶法 、資料前處理 、深度學習
外文關鍵詞： short-term solar power forecasting, energy management system, long short-term memory method, data preprocessing, deep learning
相關次數： 點閱：119  下載：9 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 台灣未來再生能源為發展主流之一，隨著再生能源在電網中佔比提高，電力系統供電可靠度會受再生能源間歇供電特性影響，勢必對電網造成諸多影響。為了提高太陽能發電預測在能源管理系統中的準確度，本論文提出一個極短期太陽能發電預測模型，藉以解決在太陽能發電案場實際使用中會遇到的問題。天氣資料使用來自Solcast每小時一筆與每五分鐘一筆的預測數據，執行太陽能發電預測系統。所提出的預測模型為長短期記憶法的雙層修正預測模型，藉由分析每筆資料特徵的關係，選擇最適合的輸入特徵，以及分析特徵輸入的先後順序，藉以提高最終預測的準確度。本文提出的太陽能發電預測模型，測試於雲林斗六友基工業區200kW以及台南沙崙智駕車33.35kW的PV案場。本論文預測每個月的小時前短期太陽能發電量預測平均誤差，上述兩個案場分別為3.6211% 和5.5214%。台南沙崙智駕車場域每五分鐘的極短期PV發電預測數據平均誤差則為2.3478%

    Renewable energy is one of Taiwan’s primary development goals. As the proportion of renewable energy in the grid increases, power system reliability can be affected by the intermittent power supply of renewable energy, adversely affecting the power grid. To improve the accuracy of solar photovoltaic (PV) power generation prediction within the energy management system, this thesis proposed a very short-term solar power generation prediction model to address the problems encountered in the actual use of solar energy. The weather data comprised hourly and every 5-min forecasting data from Solcast for use in a PV power generation forecasting system. The proposed forecasting model is a two-layer modified forecasting model of long short-term memory. By analyzing the relationship of each feature, selecting the most suitable input data, and examining the sequence of feature, the accuracy of the final forecast was improved. The forecasting model of solar power generation proposed in this thesis was tested at a 200-kW PV site in Douliu Yeoughi Industrial Park in Yunlin and a 33.35-kW PV site in Tainan Shalun Car Lab. Their average monthly PV power generation mean relative error was 3.6211% and 5.5214%, respectively. The mean relative error of the very short-term PV power generation forecast data in Shalun Car Lab every 5 minutes was 2.3478%.

    摘 要 I 誌 謝 IV ABSTRACT II CONTENTS V LIST OF FIGURES VIII LIST OF TABLES X CHAPTER 1 INTRODUCTION 1 1.1 Background and Motivation 1 1.2 Literature Review 2 1.3 Research Objective and Contribution 6 1.4 Organization of the Thesis 7 CHAPTER 2 SYSTEM DESCRIPTION AND MODELING 8 2.1 Introduction 8 2.2 Overview of Solar Energy and Photovoltaics 8 2.3 Characteristics of Machine Learning 10 2.4 Renewable Energy Forecasting 11 2.5 PV Forecasting Models 13 CHAPTER 3 THE PROPOSED FORECAST MODEL 17 3.1 The LSTM Neural Network 17 3.2 Proposed Method 19 3.3 Data Preprocessing 24 3.3.1 Datasets 24 3.3.2 Data Pre-processing and Clustering 26 3.3.3 Random Selection of Data for Training 26 3.3.4 Min–Max Normalization 27 3.3.5 Selection of Input Features with Correlation and Trend 28 CHAPTER 4 SIMULATION RESULTS 30 4.1 Introduction 30 4.2 Evaluation Indices 30 4.3 Hourly-ahead Very Short-Term Forecast Accuracy 31 4.3.1 Model Performance in Douliu, Taiwan 31 4.3.2 Model Performance in Shalun, Taiwan 37 4.4 5-Min Ahead Very Short-Term Forecast Accuracy 42 4.4.1 Model Performance in Shalun, Taiwan 42 CHAPTER 5 CONCLUSION AND FUTURE PROSPECTS 46 5.1 Conclusion 46 5.2 Future Prospects 47 REFERENCES 48

    [1] A. Catalina, C. M. Alaíz, and J. R. Dorronsoro, “Combining Numerical Weather Predictions and Satellite Data for PV Energy Nowcasting,” IEEE Transactions on Sustainable Energy, vol. 11, no. 3, pp. 1930-1937, July 2020.
    [2] Z. Zhen, F. Wang, Y. Sun, Z. Mi, C. Liu, B. Wang, and J. Lu, “SVM Based Cloud Classification Model Using Total Sky Images for PV Power Forecasting,” 2015 IEEE Power & Energy Society Innovative Smart Grid Technologies Conference (ISGT), 2015, pp. 1-5.
    [3] L. Mazorra Aguiar, B. Pereira, M. David, F. Díaz, and P. Lauret, “Use of Satellite Data to Improve Solar Radiation Forecasting with Bayesian Artificial Neural Networks,” Solar Energy, Vol 122, pp. 1309-1324, Nov. 2015.
    [4] A. Yona, T. Senjyu, T. Funabashi, and C. Kim, “Determination Method of Insolation Prediction with Fuzzy and Applying Neural Network for Long-Term Ahead PV Power Output Correction,” IEEE Transactions on Sustainable Energy, vol. 4, no. 2, pp. 527-533, April 2013.
    [5] Y. Li, J. Zhang, J. Xiao, and Y. Tan, “Short-Term Prediction of The Output Power of PV System Based on Improved Grey Prediction Model,” Proceedings of the 2014 International Conference on Advanced Mechatronic Systems, 2014, pp. 547-551.
    [6] M. J. Sanjari and H. B. Gooi, “Probabilistic Forecast of PV Power Generation Based on Higher Order Markov Chain,” IEEE Transactions on Power Systems, vol. 32, no. 4, pp. 2942-2952, July 2017.
    [7] V. Kushwaha and N. M. Pindoriya, “Very Short-Term Solar PV Generation Forecast Using SARIMA Model: A Case Study,” 2017 7th International Conference on Power Systems (ICPS), 2017, pp. 430-435.
    [8] H. Yang, C. Huang, Y. Huang, and Y. Pai, “A Weather-Based Hybrid Method for 1-Day Ahead Hourly Forecasting of PV Power Output,” IEEE Transactions on Sustainable Energy, vol. 5, no. 3, pp. 917-926, July 2014.
    [9] A. Lahouar, A. Mejri and J. Ben Hadj Slama, “Importance Based Selection Method for Day-ahead Photovoltaic Power Forecast Using Random Forests,” 2017 International Conference on Green Energy Conversion Systems (GECS), 2017, pp. 1-7.
    [10] P. Chiang, S. P. V. Chiluvuri, S. Dey, and T. Q. Nguyen, “Forecasting of Solar Photovoltaic System Power Generation Using Wavelet Decomposition and Bias-Compensated Random Forest,” 2017 Ninth Annual IEEE Green Technologies Conference (GreenTech), 2017, pp. 260-266.
    [11] G. Mosaico and M. Saviozzi, “A Hybrid Methodology for the Day-ahead PV Forecasting Exploiting a Clear Sky Model or Artificial Neural Networks,” IEEE EUROCON 2019 -18th International Conference on Smart Technologies, 2019, pp. 1-6.
    [12] Y. Sun, V. Venugopal, and A. R. Brandt, “Convolutional Neural Network for Short-Term Solar Panel Output Prediction,” 2018 IEEE 7th World Conference on Photovoltaic Energy Conversion (WCPEC) (A Joint Conference of 45th IEEE PVSC, 28th PVSEC & 34th EU PVSEC), 2018, pp. 2357-2361.
    [13] 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, 2019.
    [14] G. W. Chang and H. Lu, “Integrating Gray Data Preprocessor and Deep Belief Network for Day-Ahead PV Power Output Forecast,” IEEE Transactions on Sustainable Energy, vol. 11, no. 1, pp. 185-194, Jan. 2020.
    [15] I. J. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, and Y. Bengio, “Generative adversarial nets”, Neural Information Processing Systems, Dec. 2014.
    [16] E. Kharlova, D. May, and P. Musilek, “Forecasting Photovoltaic Power Production Using a Deep Learning Sequence to Sequence Model with Attention,” 2020 International Joint Conference on Neural Networks (IJCNN), 2020, pp. 1-7.
    [17] K. Wang, X. Qi, and H. Liu, “Photovoltaic Power Forecasting Based LSTM-Convolutional Network,'” Energy, vol. 189, p.p. 116225, Dec. 2019.
    [18] G. Li, S. Xie, B. Wang, J. Xin, Y. Li, and S. Du, “Photovoltaic Power Forecasting with a Hybrid Deep Learning Approach,” IEEE Access, vol. 8, pp. 175871-175880, 2020.
    [19] 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, 2020.
    [20] P. Li, K. Zhou, X. Lu, and S. Yang, “A Hybrid Deep Learning Model for Short-Term PV Power Forecasting,'” Applied Energy, vol. 259, February 2020.
    [21] V. Kushwaha and N. M. Pindoriya, “A SARIMA-RVFL Hybrid Model Assisted by Wavelet Decomposition for Very Short-Term Solar PV Power Generation Forecast,” Renewable Energy, Vol 140, September 2019, Pages 124-139.
    [22] J. Luffman, N. Engerer, B. King, H. Jack, M. Palmer, J. McCorkindale, H. Cutcher, and S. Elliott, “Solcast”, [Online]. Available: https://solcast.com/ [Accessed 4 July. 2021].
    [23] Coulee Ltd., “Standard Test Conditions of PV Module”, [Online]. Available: https://couleenergy.com/standard-test-conditions-of-pv-module/ [Accessed 3 June. 2021]
    [24] T. Jachja, “An Introduction to Artificial Intelligence and Machine Learning”, [Online]. Available: https://harness.io/blog/continuous-verification/ai-ml-introduction/ [Accessed 3 June. 2021].
    [25] H. Pandey, “Introduction to Artificial Intelligence, Machine Learning, Deep Learning and Data Science”, [Online]. Available: https://medium.com/analytics-vidhya/introduction-to-artificial-intelligence-machine-learning-deep-learning-and-data-science-5aa5051e9792 [Accessed 3 June. 2021].
    [26] L. Tagliaferri, “An Introduction to Machine Learning”, [Online]. Available: https://www.digitalocean.com/community/tutorials/an-introduction-to-machine-learning [Accessed 3 June. 2021].
    [27] P. Madan and S. Madhavan, “An Introduction to Deep Learning”, [Online]. Available: https://developer.ibm.com/technologies/artificial-intelligence/articles/an-introduction-to-deep-learning/ [Accessed 3 June. 2021].
    [28] 陳彥銘，太陽光電預測技術簡介， [Online]. Available: https://www.ctci.org.tw/media/6285/30-陳彥銘-太陽光電預測技術簡介.pdf [Accessed 3 June. 2021].
    [29] Introduction to Neural Networks, Advantages and Applications [Online]. Available: https://towardsdatascience.com/introduction-to-neural-networks-advantages-and-applications-96851bd1a207 [Accessed 3 June. 2021].
    [30] H. Yang, C. Huang, Y. Huang and Y. Pai, "A Weather-Based Hybrid Method for 1-Day Ahead Hourly Forecasting of PV Power Output," IEEE Transactions on Sustainable Energy, vol. 5, no. 3, pp. 917-926, July 2014.
    [31] L. Sehovac and K. Grolinger, “Deep learning for load forecasting: sequence to sequence recurrent neural networks with attention,” IEEE Access, vol. 8, pp. 36411-36426, 2020.
    [32] Kavitha S, Varuna S and Ramya R, “A comparative analysis on linear regression and support vector regression,” Online International Conference on Green Engineering and Technologies (IC-GET), 2016, pp. 1-5.
    [33] A. J. Smola and B. Schölkopf, “A tutorial on support vector regression,” Stat. Comput., vol. 14, no. 3, pp. 199–222, 2004.
    [34] C. Olah, “Understanding LSTM Networks”, [Online]. Available: https://colah.github.io/posts/2015-08-Understanding-LSTMs/ [Accessed 3 June. 2021].
    [35] Jamie M. Bright, “Solcast: Validation of a Satellite-derived Solar Irradiance Dataset,” Renewable Energy, Vol 189, 1 September 2019, Pages 435-449
    [36] H. Xu and Y. Deng, “Dependent Evidence Combination Based on Shearman Coefficient and Pearson Coefficient,” IEEE Access, vol. 6, pp. 11634-11640, 2018.
    [37] A. Senpinar, “Surface Azimuth Angle”, [Online]. Available: https://www.sciencedirect.com/topics/engineering/surface-azimuth-angle [Accessed 5 July. 2021].
    [38] M. Shi, K. Xu, J. Wang, R. Yin, T. Wang and T. Yong, “Short-Term Photovoltaic Power Forecast Based on Long Short-Term Memory Network,” International Electrical and Energy Conference (CIEEC), 2019, pp. 2110-2116.
    [39] G. Capizzi, C. Napoli and F. Bonanno, “Innovative Second-Generation Wavelets Construction With Recurrent Neural Networks for Solar Radiation Forecasting,” IEEE Transactions on Neural Networks and Learning Systems, vol. 23, no. 11, pp. 1805-1815, Nov. 2012.
    [40] A. Alfadda, R. Adhikari, M. Kuzlu and S. Rahman, “Hour-ahead solar PV power forecasting using SVR based approach,” IEEE Power & Energy Society Innovative Smart Grid Technologies Conference (ISGT), 2017, pp. 1-5.

    下載圖示 校內：2024-07-19公開
    校外：2024-07-19公開
    QR CODE