臺灣博碩士論文加值系統

智慧電網被認為是一個高度複雜的系統，其有效的管理是一項巨大的挑戰。負載預測與電力系統發展息息相關，如能源市場分析、經濟調度及安全評估，因此已被確定為如何有效管理電網的關鍵問題。如何有效利用智慧電表提供的大量資訊將是今後研究重點目標。
本文提出了三種基於長短期記憶神經網路(Long short-term memory, LSTM)的短期負載預測架構。有幾項重要改進以提升預測性能。首先，提出了融合小波轉換與長短期記憶神經網路的算法，以提高神經網路的準確度。其次，使用自相關函數篩選輸入數據以提升資料的價值，此項舉動可以有效提升機器學習效能。且結合不同特徵以提升整體架構泛用性。另外針對時間序列型資料，以不同條件進行數據分類，如此可以有效提升單一LSTM性能，且由於LSTM所需處理的資料量降低。最後提出時間序列型多步長目標預測方法，在延長預測目標距離時間之情況下，仍然可以保有一定準確度。本研究也設計了一套專用於短期負載預測的圖型使用者介面，令使用者可以簡單快速的得到所需資訊。
為了提升所提方法之效能，本文設置不同條件的案例分析，配以常用的評估指標，以驗證所提方法的可行性，並於各種角度尋求所提方法更進一步優化的可能。

The smart grid is considered to be a highly complex system, and its effective management is a huge challenge. Load forecasting is closely related to the development of power systems, such as energy market analysis, economic dispatch and safety assessment, and has therefore been identified as a key issue in how to effectively manage the grid. How to effectively use the vast amount of information provided by smart meters will be the focus of future research.
This paper proposes three short-term load forecasting architectures based on Long short-term memory neural networks (LSTM). There are several important improvements to improve forecasting performance. First, to improve the accuracy of neural networks, an algorithm combining wavelet transform and LSTM is proposed. Second, using the autocorrelation coefficient to filter the input data to enhance the value of the data, it can improve the efficiency of machine learning. And combined with different features to improve the overall architecture versatility. In addition, a time-series data segmentation architecture is proposed. For time-series data, data segmentation is performed under different conditions. This can effectively improve the performance of a single LSTM. The amount of data that LSTM needs to process is reduced. If it is combined with parallel computing technology, the required computing time can be greatly reduced. Finally, a time-series multi-step target prediction method is proposed, and the accuracy can still be maintained under the condition of prolonging the predicted target distance time.
This study also designed a graphical user interface (GUI) for short-term load forecasting, so that users can get the information quickly and easily.

目錄
誌謝 i
摘要 ii
Abstract iii
目錄 iv
圖目錄 viii
表目錄 xii
第一章 緒論 1
1.1 研究動機與背景 1
1.2 文獻回顧 2
1.3 研究目的 3
1.4 研究貢獻 3
1.5 章節內容 4
第二章 智慧電表與負載預測 7
2.1 智慧電表簡介 7
2.2 智慧電表相關議題 9
2.2.1 大數據問題 9
2.2.2 新型機器學習技術 10
2.3 負載預測 11
2.4 時間序列型資料分析 12
2.5 數據的前處理 14
2.5.1 資料清理 14
2.5.2 資料合併 15
2.5.3 資料選擇 15
2.5.4 資料濃縮 16
2.5.5 資料品質的分析 16
第三章 機器學習 17
3.1 決策樹與隨機森林 18
3.2 人工神經網路 20
3.3 遞歸神經網路 26
3.4 長短期記憶神經網路 27
3.5 過擬合問題 32
3.6 損失函數的最佳化方法 33
3.6.1 梯度下降法 33
3.6.2 RMSprop、Adam法 35
第四章 所提方法 37
4.1 數據描述 37
4.2 數據前處理 38
4.2.1 遺失及離散值處理 38
4.2.2 資料轉換 38
4.3 小波轉換 39
4.4 時間序列型特徵篩選方法 46
4.5 基因演算法 50
4.6 時間序列型單步預測架構 54
4.6.1 單步預測架構1 54
4.6.2 單步預測架構2 55
4.6.3 單步預測架構3 56
4.7 時間序列型多步長目標預測方法 57
4.7.1 方法1：遞迴式預測法 57
4.7.2 方法2：模型輸出調整 57
4.7.3 方法3：模型輸入調整 58
第五章 模擬結果與分析 59
5.1 評估指標 59
5.2 新英格蘭數據模擬 61
5.2.1 案例研究一 64
5.2.2 案例研究二 67
5.2.3 案例研究三 72
5.2.4 案例研究四 76
5.2.5 案例研究五 77
5.2.6 案例研究六 82
5.2.7 案例研究七 84
5.3 台電智慧電表數據模擬 86
5.3.1 案例分析一 89
5.3.2 案例分析二 95
5.3.3 案例分析三 96
5.3.4 案例分析四 106
第六章 結論與未來工作 107
6.1 結論 107
6.2 未來工作 108
參考文獻 109
附錄 115
A.1 圖形使用者介面 115
A.2 負載預測操作說明 116
A.3 輸入資料分析 121

[1]D. W. Bunn and E. D. Farmer, ComparativeModels for Electrical LoadForecasting. New York, NY, USA: Wiley, 1985, Chichester.
[2]C. Nataraja, M. Gorawar, G. Shilpa, and J. S. Harsha, “Short term load forecasting using time series analysis: A case study for karnataka, india,” Inte. Jour. of Engi. Scie. and Inno. Tech., vol. 1, pp. 45–53, 2012.
[3]X. Cao, S. Dong, Z. Wu, and Y. Jing, "A data-driven hybrid optimization model for short-term residential load forecasting," Computer and Information Technology; Ubiquitous Computing and Communications; Dependable, Autonomic and Secure Computing; Pervasive Intelligence and Computing (CIT/IUCC/DASC/PICOM), 2015 IEEE International Conference on, 2015, pp. 283-287.
[4]L. H. Tsoukalas and R. Gao, “From smart grids to an energy internet:Assumptions, architectures and requirements,” in Proc. 3rd Int. Conf.Electr. Utility Deregulation Restructuring Power Technol., 2008, pp.94–98.
[5]D.Xie, J.Xu, and J.Yu, “The physical series algorithm of mid-long term load forecasting,” Electr. Power Syst. Res., vol. 53, no. 1, pp. 31–37,2000.
[6]G. J.Tsekouras, N. D. Hatziargiriou, and E.N.Dialynas, “An optimized adaptive neural network for annual midterm energy forecasting,” IEEE Trans. Power Syst., vol. 21, no. 1, pp. 385–391, Feb. 2006.
[7]A. G. Bakirtzis, J. B. Theoharis, S. J. Kiartzis, and K. J. Satsios, “Short-term load forecasting using fuzzy neural networks,” IEEE Trans. Power Syst., vol. 10, no. 3, pp. 1518–1524, Aug. 1995.
[8]K. Liu, S. Subbarayan, R. R. Shoults, M. T. Manry, C. Kwan, and F.L. Lewis et al., “Comparison of very short-term load forecasting techniques,” IEEE Trans. Power Syst., vol. 11, no. 2, pp. 877–882, May1996.
[9]S. J. Huang and K. R. Shih, “Short-term load forecasting via ARIMA model identification including non-Gaussian process considerations,” IEEE Trans. Power Syst., vol. 18, no. 2, pp. 673–679, May 2003.
[10]T. Senjyu, H. Takara, K. Uezato, and T. Funabashi, “One-hour-ahead load forecasting using neural network,” IEEE Trans. Power Syst., vol.17, no. 1, pp. 113–118, Feb. 2002.
[11]P. K. Dash, A. C. Liew, and S. Rahman, “Fuzzy neural network and fuzzy expert system for load forecasting,” Proc. Inst. Electr.Eng.—Gener., Transm. Distrib., vol. 143, pp. 106–114, 1996.
[12]S. Fan and L. Chen, “Short-term load forecasting based on an adaptive hybrid method,” IEEE Trans. Power Syst., vol. 21, no. 1, pp. 392–401,Feb. 2006.
[13]K. Liu, K. S. Subbarayan, R. R. Shoults, M. T. Manry, C. Kwan, F.L. Lewis, and J. Naccarino, “Comparison of very short-term load forecasting techniques,” IEEE Trans. Power Syst., vol. 11, no. 2, pp.877–882, May 1996.
[14]X. Chen, Z. Y. Dong, K. Meng, Y. Xu, K. P. Wong, and H. W. Ngan,“Electricity price forecasting with extreme learning machine and bootstrapping,” IEEE Trans. Power Syst., vol. 27, no. 4, pp. 2055–2062,Nov. 2012.
[15]R. Zhang, Z. Y. Dong, Y. Xu, K. Meng, and K. P. Wong, “Short-term load forecasting of Australian national electricity market by an ensemble model of extreme learning machine,” IET Gener., Trans. Distrib.,vol. 7, pp. 391–397, 2013.
[16]S. Li, P. Wang, and L. Goel, “Short-term load forecasting by wavelet transform and evolutionary extreme learning machine,” Electr. PowerSyst. Res., vol. 122, pp. 96–103, 2015.
[17]P. Shamsollahi, K. W. Cheung, Q. Chen, and E. H. Germain, “A neural network based very short term load forecaster for the interim ISO newEngland electricity market,” in Proc. IEEE Power Ind. Comput. Appl. Conf., 2001, pp. 217–222.
[18]W. Charytoniuk and M. S. Chen, “Very short-term load forecasting using artificial neural networks,” IEEE Trans. Power Syst., vol. 15, no.1, pp. 263–268, Feb. 2000.
[19]L. C. M. de Andrade and I. N. da Silva, “Very short-term load forecasting using a hybrid neuro-fuzzy approach,” in Proc. 11th BrazilianSymp. Neural Network, 2010, pp. 115–120.
[20]L. C. M. de Andrade and I. N. da Silva, “Using intelligent system approach for very short-term load forecasting purposes,” in Proc. IEEE Int. Energy Conf., 2010, pp. 694–699.
[21]Q. Pang and M. Zhang, “Very short-term load forecasting based on neural network and rough set,” in Proc. Int. Conf. Intell. Comp. Tech. Autom., 2010, pp. 1132–1135.
[22]C. Guan, P. B. Luh, M. A. Coolbeth, Y. Zhao, L. D.Michel, Y. Chen, C. J. Manville, P. B. Friedland, and S. J. Rourke, “Very short-term load forecasting: Multilevel wavelet neural networks with data pre-filtering,” in Proc. IEEE Power & Energy Soc. General Meeting, 2009, pp. 1–8.
[23]H. Yang, H. Ye, G. Wang, and T. Hu, “Fuzzy neural very-short-term load forecasting based on chaotic dynamics reconstruction,” Chaos Solitons Fractals, vol. 29, no. 2, pp. 462–469, 2006.
[24]S. Kawauchi, H. Sugihara, and H. Sasaki, “Development of very-shortterm load forecasting based on chaos theory,” Trans. Inst. Electr. Eng. Jpn. B, vol. 123, no. 5, pp. 646–653, 2004.
[25]C.Guan, P. B. Luh, L.D.Michel, M. A. Coolbeth, and P. B. Friedland, “Hybrid Kalman algorithms for very short-term load forecasting and confidence interval estimation,” in Proc. IEEE Power & Energy Soc. General Meeting, 2010, pp. 1–8.
[26]A. Setiawan, I. Koprinska, and V. Aggelidis, “Very short-term electricity load demand forecasting using support vector regression,” in Proc. Int. Joint Conf. Neural Networks, 2009, pp. 2888–2894.
[27]W. C. Hong, “Electric load forecasting by seasonal recurrent SVR (support vector regression) with chaotic artificial bee colony algorithm,” Energy, vol. 36, pp. 5568–5578, 2011.
[28]Y. M. Wi, S. K. Joo, and K. B. Song, “Holiday load forecasting using fuzzy polynomial regression with weather feature selection and adjustment,” IEEE Trans. Power Syst., vol. 27, no. 2, pp. 596–603, May 2012.
[29]Department for Business Energy and Industrial Strategy, “Smart Meters,Quarterly Report to end December 2016, Great Britain,” Tech.Rep., 2017.
[30]“How many smart meters are installed in the united states, and who has them?” https://www.eia.gov/tools/faqs/faq.php?id=108&t=3, accessed July 31, 2017.
[31]“Number of electric smart meters deployed in the u.s. from 2008 to 2016,” https://www.statista.com/statistics/499704/ number-of-smart-meters-in-the-united-states/, accessed July 31, 2017.
[32]R. R. Mohassel, A. Fung, F. Mohammadi, and K. Raahemifar, “A survey on advanced metering infrastructure,” International Journal of Electrical Power & Energy Systems, vol. 63, pp. 473–484, 2014.
[33]J. Yang, J. Zhao, F. Luo, F. Wen, and Z. Y. Dong, “Decision-making for electricity retailers: A brief survey,” IEEE Trans. Smart Grid, vol. PP, no. 99, pp. 1–1, 2017.
[34]National Science Foundation, “Smart grids big data.” https://www.nsf.gov/awardsearch/showAward?AWD ID=1636772& HistoricalAwards=false, 2016.
[35]X. Liu, A. Heller, and P. S. Nielsen, “CITIESData: a smart city data management framework,” Knowledge and Information Systems, vol. 53, no. 3, pp. 699–722, 2017.
[36]“Bits to energy lab projects,” http://www.bitstoenergy.ch/home/ projects/, accessed July 31, 2017.
[37]Siebel Energy Institute, “Advancing the science of smart energy.” http: //www.siebelenergyinstitute.org/, 2016.
[38]“Wp3 overview,” https://webgate.ec.europa.eu/fpfis/mwikis/ essnetbigdata/index.php/WP3 overview, accessed July 31, 2017.
[39]SAS, “Utility analytics in 2017: Aligning data and analytics with business strategy,” Tech. Rep., 2017.
[40]T. Hong, D. Gao, T. Laing, D. Kruchten, and J. Calzada, “Training energy data scientists: universities and industry need to work together to bridge the talent gap,” IEEE Power and Energy Magazine, vol. 16, no. 3, pp. –, May 2018.
[41]T. Hong, C. Chen, J. Huang, N. Lu, L. Xie, and H. Zareipour, “Guest editorial big data analytics for grid modernization,” IEEE Trans. Smart Grid, vol. 7, no. 5, pp. 2395–2396, 2016.
[42]T. Hong, “Big data analytics: making smart grid smarter,” IEEE Power and Energy Magazine, vol. 16, no. 3, pp. –, May 2018.
[43]Z. Lv, H. Song, P. Basanta-Val, A. Steed, and M. Jo, “Next-generation big data analytics: State of the art, challenges, and future research topics,” IEEE Trans. Industrial Informatics, vol. 13, no. 4, pp. 1891–1899, Aug 2017.
[44]J. Baek, Q. H. Vu, J. K. Liu, X. Huang, and Y. Xiang, “A secure cloudcomputing based framework for big data information management of smart grid,” IEEE Trans. cloud computing, vol. 3, no. 2, pp. 233–244,2015.
[45]S. Bera, S. Misra, and J. J. Rodrigues, “Cloud computing applicationsfor smart grid: A survey,” IEEE Trans. Parallel and Distributed Systems, vol. 26, no. 5, pp. 1477–1494, 2015.
[46]S. Mittal, “A survey of techniques for architecting and managing gpu register file,” IEEE Trans. Parallel and Distributed Systems, vol. 28, no. 1, pp. 16–28, 2017.
[47]A. Rodriguez and A. Laio, “Clustering by fast search and find of density peaks,” Science, vol. 344, no. 6191, pp. 1492–1496, 2014.
[48]Y. Wang, Q. Chen, C. Kang, and Q. Xia, “Clustering of electricity consumption behavior dynamics toward big data applications,” IEEE Trans. Smart Grid, vol. 7, no. 5, pp. 2437–2447, 2016.
[49]F. A. Gers, J. Schmidhuber, and F. Cummins, “Learning to forget: Continual prediction with lstm,” Neural computation, vol. 12, no. 10, pp. 2451–2471, 2000.
[50]D. L. Marino, K. Amarasinghe, and M. Manic, “Building energy load forecasting using deep neural networks,” in Industrial Electronics Society, IECON 2016-42nd Annual Conference of the IEEE. IEEE, 2016, pp. 7046–7051.
[51]S. J. Pan and Q. Yang, “A survey on transfer learning,” IEEE Trans. knowledge and data engineering, vol. 22, no. 10, pp. 1345–1359, 2010.
[52]Y. Bengio, “Deep learning of representations for unsupervised and transfer learning,” in Proceedings of ICML Workshop on Unsupervised and Transfer Learning, 2012, pp. 17–36.
[53]T. Diethe and M. Girolami, “Online learning with (multiple) kernels: A review,” Neural computation, vol. 25, no. 3, pp. 567–625, 2013.
[54]Y. Xie, W. Zhang, C. Li, S. Lin, Y. Qu, and Y. Zhang, “Discriminative object tracking via sparse representation and online dictionary learning,” IEEE Trans. cybernetics, vol. 44, no. 4, pp. 539–553, 2014.
[55]Q. Zhang, C. Zhu, L. T. Yang, Z. Chen, L. Zhao, and P. Li, “An Incremental CFS Algorithm for Clustering Large Data in Industrial Internet of Things,” IEEE Trans. Industrial Informatics, vol. 13, no. 3, pp. 1193–1201, 2017.
[56]Z. C. Lipton, J. Berkowitz, and C. Elkan, “A critical review of recurrent neural networks for sequence learning,” arXiv preprint arXiv:1506.00019, 2015.
[57]I. Sutskever, O. Vinyals, and Q. V. Le, “Sequence to sequence learning with neural networks,” in Advances in neural information processing systems, 2014, pp. 3104–3112.
[58]Y. Bengio, P. Simard, and P. Frasconi, “Learning long-term dependencies with gradient descent is difficult,” IEEE tran. on neur. netw., vol. 5, no. 2, pp. 157–166, 1994.
[59]S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neur. comp., vol. 9, no. 8, pp. 1735–1780, 1997.
[60]N. Amjady and F. Keynia, “Short-term load forecasting of power systems by combination of wavelet transform and neuro-evolutionary algorithm,” Energy, vol. 34, pp. 46–57, 2009.
[61]A. J. R. Reis and A. P. A. da Silva, “Feature extraction via multiresolution analysis for short-term load forecasting,” IEEE Trans. Power Syst., vol. 20, no. 1, pp. 189–198, Feb. 2005.