準確的負荷預測系統對於電力公司來說是有效管理系統運行和調度所必需的。然而，面對公眾假期，由於人類行為的急劇變化，負荷消耗模式的行為無法預測。因此，假期負荷預測是一個具有挑戰性的問題，因為假期期間的預測準確性與一般的工作日/週末相比有所下降。在本文中，我們提出了DynaLSTM，這是一種混合型短期假期預測框架，旨在預測即將到來的公共假期之前的最近一個工作日的公共假期的負荷消耗。 DynaLSTM由動態時間規整（DTW）和長期短期記憶（LSTM）組成。 DTW用於預測最近工作日和任何先前的補償性假期（如果有）的負荷消耗，而LSTM用於預測目標公共假日的高度不可預測的負荷消耗。此外，我們還提出了一種數據選擇程序（DSP），以系統地選擇訓練數據來訓練LSTM模型。當前結果表明，與其他方法相比，我們提出的框架在不同長度的假期中表現更好。

An accurate load forecasting system is necessary for a power company to efficiently manage system operation and scheduling. In the face of a public holiday, however, the load consumption pattern behaves unpredictably as a result of the drastic change in human behavior. Thus, holiday load forecasting is a challenging issue because the forecasting accuracy in holiday period fell compared to general weekdays/weekends. In this thesis, we propose DynaLSTM, which is a hybrid short-term holiday forecasting framework designed to predict the load consumption of a public holiday on the nearest workday prior to the incoming public holiday. DynaLSTM consists of dynamic time warping (DTW) and long-short term memory (LSTM). DTW is used to predict the load consumption of the nearest workday and any preceding compensatory holiday(s), if any, while LSTM is used to predict the highly unpredictable load consumption of the target public holiday. Additionally, we also propose a data selection procedure (DSP) to systematically select training data to train the LSTM model with. Present results demonstrate that our proposed framework performs better across holidays of different length compared to other methods.

[1] M. A. Ortega-Vazquez and D. S. Kirschen, \Economic impact assessment of load fore-
cast errors considering the cost of interruptions," in 2006 IEEE Power Engineering
Society General Meeting, pp. 1{8, 2006.
[2] B. F. Hobbs, S. Jitprapaikulsarn, S. Konda, V. Chankong, K. A. Loparo, and D. J.
Maratukulam, \Analysis of the value for unit commitment of improved load fore-
casts," IEEE Trans. Power Syst., vol. 14, no. 4, pp. 1342{1348, 1999.
[3] A. Almalaq and J. J. Zhang, Deep learning application: load forecasting in big data of
smart grids, vol. 865, pp. 103{128. Cham: Springer International Publishing, 2020.
[4] J. Zheng, C. Xu, Z. Zhang, and X. Li, \Electric load forecasting in smart grids using
long-short-term-memory based recurrent neural network," in 2017 51st Annu. Conf.
Information Sciences and Systems, pp. 1{6, 2017.
[5] P. Zeng, C. Sheng, and M. Jin, \A learning framework based on weighted knowledge
transfer for holiday load forecasting," J. Mod. Power Syst. Clean Energy, vol. 7,
pp. 1{21, Mar 2019.
[6] G. Forestier, F. Petitjean, H. A. Dau, G. I. Webb, and E. Keogh, \Generating syn-
thetic time series to augment sparse datasets," in 2017 IEEE International Conference
on Data Mining (ICDM), pp. 865{870, 2017.
[7] Y.-M. Wi, S.-K. Joo, and K.-B. Song, \Holiday load forecasting using fuzzy polyno-
mial regression with weather feature selection and adjustment," IEEE Trans. Power
Syst., vol. 27, pp. 596{603, May 2012.
[8] Q. Ding, H. Zhang, T. Huang, and J. Zhang, \A holiday short term load forecasting
considering weather information," in 2005 International Power Engineering Confer-
ence, pp. 1{61, 2005.
[9] Y. Cai, Q. Xie, C. Wang, and F. Lu, \Short-term load forecasting for city holidays
based on genetic support vector machines," in 2011 International Conference on
Electrical and Control Engineering, pp. 3144{3147, 2011.
[10] R. S. S. Fernandes, Y. K. Bichpuriya, M. S. S. Rao, and S. A. Soman, \Day ahead load
forecasting models for holidays in indian context," in 2011 International Conference
on Power and Energy Systems, pp. 1{5, 2011.
[11] F. Ziel, \Modeling public holidays in load forecasting: a german case study," J. Mod.
Power Syst. Clean Energy, vol. 6, pp. 191{207, Mar 2018.
[12] S. Fallah, M. Ganjkhani, S. Shamshirband, and K.-w. Chau, \Computational intelli-
gence on short-term load forecasting: a methodological overview," Energies, vol. 12,
pp. 1{21, Jan 2019.
[13] J. Y. Hong, S. H. Park, and J.-G. Baek, \Segmented dynamic time warping based sig-
nal pattern classication," in 2019 IEEE International Conference on Computational
Science and Engineering (CSE) and IEEE International Conference on Embedded
and Ubiquitous Computing (EUC), pp. 263{265, 2019.
[14] M. Brown and L. Rabiner, \Dynamic time warping for isolated word recognition
based on ordered graph searching techniques," in ICASSP '82. IEEE International
Conference on Acoustics, Speech, and Signal Processing, vol. 7, pp. 1255{1258, 1982.
[15] T. Teeraratkul, D. ONeill, and S. Lall, \Shape-based approach to household elec-
tric load curve clustering and prediction," IEEE Trans. Smart Grid, vol. 9, no. 5,
pp. 5196{5206, 2018.
[16] A. Ahmad, N. Javaid, A. Mateen, M. Awais, and Z. A. Khan, \Short-term load
forecasting in smart grids: an intelligent modular approach," Energies, vol. 12, pp. 1{
21, Jan 2019.
[17] G. Dudek, \Short-term load cross-forecasting using pattern-based neural models," in
2015 16th International Scientic Conference on Electric Power Engineering (EPE),
pp. 179{183, 2015.
[18] K. Chen, K. Chen, Q. Wang, Z. He, J. Hu, and J. He, \Short-term load forecasting
with deep residual networks," IEEE Trans. Smart Grid, vol. 10, no. 4, pp. 3943{3952,
2019.
[19] W. Kong, Z. Y. Dong, Y. Jia, D. J. Hill, Y. Xu, and Y. Zhang, \Short-term residential
load forecasting based on lstm recurrent neural network," IEEE Trans. Smart Grid,
vol. 10, no. 1, pp. 841{851, 2019.
[20] M. Tan, S. Yuan, S. Li, Y. Su, H. Li, and F. He, \Ultra-short-term industrial power
demand forecasting using lstm based hybrid ensemble learning," IEEE Trans. Power
Syst., vol. 35, no. 4, pp. 2937{2948, 2020.
[21] M. S. Hossain and H. Mahmood, \Short-term load forecasting using an lstm neural
network," in 2020 IEEE Power and Energy Conference at Illinois (PECI), pp. 1{6,
2020.
[22] L. Tang, Y. Yi, and Y. Peng, \An ensemble deep learning model for short-term
load forecasting based on arima and lstm," in 2019 IEEE International Conference
on Communications, Control, and Computing Technologies for Smart Grids (Smart-
GridComm), pp. 1{6, 2019.
[23] R. Asadi and A. C. Regan, \A spatio-temporal decomposition based deep neural
network for time series forecasting," Applied Soft Computing, vol. 87, pp. 1{13, Feb
2020.
[24] Z. Yu, Z. Niu, W. Tang, and Q.Wu, \Deep learning for daily peak load forecastinga
novel gated recurrent neural network combining dynamic time warping," IEEE Ac-
cess, vol. 7, pp. 17184{17194, Jan 2019.
[25] S. Theodoridis and K. Koutroumbas, Pattern recognition, pp. 324{325. San Diego,
CA, USA: Academic Press, 2009.
[26] N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov,
\Dropout: a simple way to prevent neural networks from overtting," J. Mach. Learn.
Res., vol. 15, pp. 1929{1958, Jan 2014.
[27] F. Ziel and R. Weron, \Day-ahead electricity price forecasting with high-dimensional
structures: univariate vs. multivariate modeling frameworks," Energy Economics,
vol. 70, pp. 396{420, Feb 2018.
[28] F. Chollet et al., \Keras." https://keras.io, 2015.
[29] S. Liu, H. Ji, and M. C. Wang, \Nonpooling convolutional neural network forecasting
for seasonal time series with trends," IEEE Trans. Neural Netw. Learn. Syst., vol. 31,
no. 8, pp. 2879{2888, 2020.
[30] S. A. Soliman and A. M. Alkandari, Electrical load forecasting: modeling and model
construction, p. 86. Burlington, MA, USA: Butterworth-Heinemann, 2010.
[31] A. RochaReis and A. AlvesdaSilva, \Feature extraction via multiresolution analysis
for short-term load forecasting," IEEE Trans. Power Syst., vol. 20, pp. 189{198, Feb
2005.