近年來地球資源消耗日益嚴重，化石燃料仍然是全球最廣泛使用且最重要的發電來源。然而化石燃料消耗的速度比再生能源的速度還要快，同時也是地球污染和溫室氣體排放的主要來源，不但對人類的健康產生危害，全球暖化效應也急遽增加。根據國際能源總署 (The International Energy Agency, IEA) 在2020年世界能源展望報告中預測，2020年至2030年全球再生能源電力需求將提高至三分之二，約增加全球電力需求的80%。近年來，光伏時代的普及，使得太陽能在電力系統中的比重迅速增加，光電發電利用太陽輻射來轉換電能，因此光電系統的性能直接受太陽輻照度的影響。為了穩定發電，準確預測太陽輻照度變得重要。
本項研究通過Solcast網站獲取2021到2022年太陽輻照量進行三個實驗。第一種是透過自迴歸積分移動平均 (Autoregressive Integrated Moving Average, ARIMA)、自迴歸條件異方差（Autoregressive Conditional Heteroskedasticity, ARCH）、廣義自迴歸條件異方差（Generalized Autoregressive Conditional Heteroskedasticity, GARCH）等統計模型以及長短期記憶（Long Short Term Memory, LSTM）、雙向長短期記憶(Bidirectional Long Short Term Memory, Bi-LSTM) 和門控循環單元(Gated Recurrent Unit, GRU)等深度學習模型來預測太陽輻照量。第二種是將信號預測分解技術與深度學習模型結合，以經驗模態分解（Empirical Mode Decomposition, EMD）、增強經驗模態分解（Ensemble Empirical Mode Decomposition, EEMD）和自適應雜訊完全整合經驗模態分解（Complete Ensemble Empirical Mode Decomposition With Adaptive Noise, CEEMDAN）三種信號分解方法分解，然後進行深度學習訓練與預測。最後是使用混合深度學習模型LSTM-Bi-LSTM、LSTM-GRU、Bi-LSTM-LSTM、Bi-LSTM-GRU、GRU-LSTM和GRU-Bi-LSTM模型進行訓練與預測。結果顯示，採用信號分解方法搭配混合深度學習模型進行預測可以提高預測精度，本研究以EEMD-GRU-Bi-LSTM能最精準的預測太陽輻照量。

In recent years, the popularity of the photovoltaic (PV) era has caused the proportion of solar energy in the power system to increase rapidly. The overall performance and efficacy of a photovoltaic structure is affected by a variety of factors, including solar irradiance, climatic conditions, temperature, pollution and obstructions, and the tilt angle of the photovoltaic panels. Photovoltaic power generation uses solar radiation to convert electrical energy, so the performance of the photovoltaic system is directly affected by solar irradiance. In order to stabilize power generation, accurate prediction of solar irradiance becomes increasingly important.
This study conducted three experiments. The first is to predict the amplitude and illuminance through statistical models such as Autoregressive Integrated Moving Average (ARIMA), Autoregressive Conditional Heteroskedasticity (ARCH), Generalized Autoregressive Conditional Heteroskedasticity (GARCH) and deep learning models such as Long Short-Term Memory (LSTM), Bidirectional LSTM (Bi-LSTM) and Gated Recurrent Unit (GRU). The second is to predict the signal The decomposition technology is combined with the deep learning model. The signal data is first decomposed through Empirical Mode Decomposition (EMD), Enhanced EMD (EEMD) and Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), and then put into deep learning in the model. Last is use hybrid deep learning models LSTM-Bi-LSTM, LSTM-GRU, Bi-LSTM-LSTM, Bi-LSTM- GRU, GRU-LSTM and GRU-Bi-LSTM to predict. The results show that using signal decomposition method for prediction can improve the prediction accuracy, and EEMD-GRU-LSTM can predict solar irradiance more accurately.

[1]. Wang, L., Mao, M., Xie, J., Liao, Z., Zhang, H., & Li, H. (2023). Accurate solar PV power prediction interval method based on frequency-domain decomposition and LSTM model. Energy, 262, 125592.
[2]. Kumari, P., & Toshniwal, D. (2021). Long short term memory–convolutional neural network based deep hybrid approach for solar irradiance forecasting. Applied Energy, 295, 117061.
[3]. Huynh, A. N. L., Deo, R. C., Ali, M., Abdulla, S., & Raj, N. (2021). Novel short-term solar radiation hybrid model: Long short-term memory network integrated with robust local mean decomposition. Applied Energy, 298, 117193.
[4]. Sehrawat, N., Vashisht, S., & Singh, A. (2023). Solar irradiance forecasting models using machine learning techniques and digital twin: A case study with comparison. International Journal of Intelligent Networks, 4, 90-102.
[5]. Hu, L., Cao, Y., & Yin, L. (2024). Fractional-order long-term price guidance mechanism based on bidirectional prediction with attention mechanism for electric vehicle charging. Energy, 130639.
[6]. Song, M., Cheng, L., Du, M., Sun, C., Ma, J., & Ge, H. (2023). Charging station location problem for maximizing the space-time-electricity accessibility: A Lagrangian relaxation-based decomposition scheme. Expert Systems with Applications, 222, 119801.
[7]. Salvarli, M. S., & Salvarli, H. (2020). For sustainable development: Future trends in renewable energy and enabling technologies. In Renewable energy-resources, challenges and applications. IntechOpen.
[8]. Yu, M., Niu, D., Gao, T., Wang, K., Sun, L., Li, M., & Xu, X. (2023). A novel framework for ultra-short-term interval wind power prediction based on RF-WOA-VMD and BiGRU optimized by the attention mechanism. Energy, 269, 126738.
[9]. Perez, R., Zweibel, K., & Hoff, T. E. (2011). Solar power generation in the US: Too expensive, or a bargain?. Energy Policy, 39(11), 7290-7297.
[10]. Li, L. L., Wen, S. Y., Tseng, M. L., & Wang, C. S. (2019). Renewable energy prediction: A novel short-term prediction model of photovoltaic output power. Journal of Cleaner Production, 228, 359-375.
[11]. Li, G., Wei, X., & Yang, H. (2023). Decomposition integration and error correction method for photovoltaic power forecasting. Measurement, 208, 112462.
[12]. Jailani, N. L. M., Dhanasegaran, J. K., Alkawsi, G., Alkahtani, A. A., Phing, C. C., Baashar, Y., ... & Tiong, S. K. (2023). Investigating the power of LSTM-based models in solar energy forecasting. Processes, 11(5), 1382.
[13]. Jungbluth, N., Stucki, M., Frischknecht, R., & Buesser, S. (2009). Photovoltaics. Sachbilanzen von Energiesystemen: Grundlagen für den ökologischen Vergleich von Energiesystemen und den Einbezug von Energiesystemen in Ökobilanzen für die Schweiz. ecoinvent report, (6-XII), 16-69.
[14]. Djaafari, A., Ibrahim, A., Bailek, N., Bouchouicha, K., Hassan, M. A., Kuriqi, A., ... & El-Kenawy, E. S. M. (2022). Hourly predictions of direct normal irradiation using an innovative hybrid LSTM model for concentrating solar power projects in hyper-arid regions. Energy Reports, 8, 15548-15562.
[15]. Li, Q., Zhang, D., & Yan, K. (2023). A solar irradiance forecasting framework based on the CEE-WGAN-LSTM model. Sensors, 23(5), 2799.
[16]. Liu, Z. F., Luo, S. F., Tseng, M. L., Liu, H. M., Li, L., & Mashud, A. H. M. (2021). Short-term photovoltaic power prediction on modal reconstruction: A novel hybrid model approach. Sustainable Energy Technologies and Assessments, 45, 101048.
[17]. Lima, M. A. F., Carvalho, P. C., Fernández-Ramírez, L. M., & Braga, A. P. (2020). Improving solar forecasting using Deep Learning and Portfolio Theory integration. Energy, 195, 117016.
[18]. Tercha, W., Tadjer, S. A., Chekired, F., & Canale, L. (2024). Machine Learning-Based Forecasting of Temperature and Solar Irradiance for Photovoltaic Systems. Energies, 17(5), 1124.
[19]. Sharma, A., & Kakkar, A. (2018). Forecasting daily global solar irradiance generation using machine learning. Renewable and Sustainable Energy Reviews, 82, 2254-2269.
[20]. Benali, L., Notton, G., Fouilloy, A., Voyant, C., & Dizene, R. (2019). Solar radiation forecasting using artificial neural network and random forest methods: Application to normal beam, horizontal diffuse and global components. Renewable energy, 132, 871-884.
[21]. Fan, J., Wu, L., Zhang, F., Cai, H., Zeng, W., Wang, X., & Zou, H. (2019). Empirical and machine learning models for predicting daily global solar radiation from sunshine duration: A review and case study in China. Renewable and Sustainable Energy Reviews, 100, 186-212.
[22]. Hissou, H., Benkirane, S., Guezzaz, A., Azrour, M., & Beni-Hssane, A. (2023). A novel machine learning approach for solar radiation estimation. Sustainability, 15(13), 10609.
[23]. Demir, V., & Citakoglu, H. (2023). Forecasting of solar radiation using different machine learning approaches. Neural Computing and Applications, 35(1), 887-906.
[24]. Chodakowska, E., Nazarko, J., Nazarko, Ł., Rabayah, H. S., Abendeh, R. M., & Alawneh, R. (2023). Arima models in solar radiation forecasting in different geographic locations. Energies, 16(13), 5029.
[25]. David, M., Ramahatana, F., Trombe, P. J., & Lauret, P. (2016). Probabilistic forecasting of the solar irradiance with recursive ARMA and GARCH models. Solar Energy, 133, 55-72.
[26]. Alkabbani, H., Ahmadian, A., Zhu, Q., & Elkamel, A. (2021). Machine learning and metaheuristic methods for renewable power forecasting: a recent review. Frontiers in Chemical Engineering, 3, 665415.
[27]. Hinton, G. E., Osindero, S., & Teh, Y. W. (2006). A fast learning algorithm for deep belief nets. Neural computation, 18(7), 1527-1554.
[28]. Pazikadin, A. R., Rifai, D., Ali, K., Malik, M. Z., Abdalla, A. N., & Faraj, M. A. (2020). Solar irradiance measurement instrumentation and power solar generation forecasting based on Artificial Neural Networks (ANN): A review of five years research trend. Science of The Total Environment, 715, 136848.
[29]. Jaihuni, M., Basak, J. K., Khan, F., Okyere, F. G., Sihalath, T., Bhujel, A., ... & Kim, H. T. (2022). A novel recurrent neural network approach in forecasting short term solar irradiance. ISA transactions, 121, 63-74.
[30]. Qing, X., & Niu, Y. (2018). Hourly day-ahead solar irradiance prediction using weather forecasts by LSTM. Energy, 148, 461-468.
[31]. Marinho, F. P., Rocha, P. A., Neto, A. R., & Bezerra, F. D. (2023). Short-term solar irradiance forecasting using CNN-1D, LSTM, and CNN-LSTM deep neural networks: A case study with the Folsom (USA) dataset. Journal of Solar Energy Engineering, 145(4), 041002.
[32]. Yan, K., Shen, H., Wang, L., Zhou, H., Xu, M., & Mo, Y. (2020). Short-term solar irradiance forecasting based on a hybrid deep learning methodology. Information, 11(1), 32.
[33]. Huang, X., Li, Q., Tai, Y., Chen, Z., Zhang, J., Shi, J., ... & Liu, W. (2021). Hybrid deep neural model for hourly solar irradiance forecasting. Renewable Energy, 171, 1041-1060.
[34]. Huang, N. E., Shen, Z., Long, S. R., Wu, M. C., Shih, H. H., Zheng, Q., ... & Liu, H. H. (1998). The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis. Proceedings of the Royal Society of London. Series A: mathematical, physical and engineering sciences, 454(1971), 903-995.
[35]. Wu, Z., & Huang, N. E. (2009). Ensemble empirical mode decomposition: a noise-assisted data analysis method. Advances in adaptive data analysis, 1(01), 1-41.
[36]. Torres, M. E., Colominas, M. A., Schlotthauer, G., & Flandrin, P. (2011, May). A complete ensemble empirical mode decomposition with adaptive noise. In 2011 IEEE international conference on acoustics, speech and signal processing (ICASSP) (pp. 4144-4147). IEEE.
[37]. Graves, A., & Graves, A. (2012). Long short-term memory. Supervised sequence labelling with recurrent neural networks, 37-45.
[38]. Zazo, R., Lozano-Diez, A., Gonzalez-Dominguez, J., T. Toledano, D., & Gonzalez-Rodriguez, J. (2016). Language identification in short utterances using long short-term memory (LSTM) recurrent neural networks. PloS one, 11(1), e0146917.
[39]. Li, L., & Zhang, T. (2021). Research on text generation based on lstm. International Core Journal of Engineering, 7(5), 525-535.
[40]. Graves, A., Jaitly, N., & Mohamed, A. R. (2013, December). Hybrid speech recognition with deep bidirectional LSTM. In 2013 IEEE workshop on automatic speech recognition and understanding (pp. 273-278). IEEE.
[41]. Zhang, X., Liang, X., Zhiyuli, A., Zhang, S., Xu, R., & Wu, B. (2019, July). At-lstm: An attention-based lstm model for financial time series prediction. In IOP Conference Series: Materials Science and Engineering (Vol. 569, No. 5, p. 052037). IOP Publishing.
[42]. Karevan, Z., & Suykens, J. A. (2020). Transductive LSTM for time-series prediction: An application to weather forecasting. Neural Networks, 125, 1-9.
[43]. Chang, C. W., Chang, C. Y., & Lin, Y. Y. (2022). A hybrid CNN and LSTM-based deep learning model for abnormal behavior detection. Multimedia Tools and Applications, 81(9), 11825-11843.
[44]. Liu, Q., Zhou, F., Hang, R., & Yuan, X. (2017). Bidirectional-convolutional LSTM based spectral-spatial feature learning for hyperspectral image classification. Remote Sensing, 9(12), 1330.
[45]. Kosko, B. (1988). Bidirectional associative memories. IEEE Transactions on Systems, man, and Cybernetics, 18(1), 49-60.
[46]. Su, Y., & Kuo, C. C. J. (2019). On extended long short-term memory and dependent bidirectional recurrent neural network. Neurocomputing, 356, 151-161.
[47]. Chen, P., Xiao, Q., Zhang, J., Xie, C., & Wang, B. (2020). Occurrence prediction of cotton pests and diseases by bidirectional long short-term memory networks with climate and atmosphere circulation. Computers and Electronics in Agriculture, 176, 105612.
[48]. Zhu, J., Sun, K., Jia, S., Lin, W., Hou, X., Liu, B., & Qiu, G. (2018). Bidirectional long short-term memory network for vehicle behavior recognition. Remote Sensing, 10(6), 887.
[49]. Chung, J., Gulcehre, C., Cho, K., & Bengio, Y. (2014). Empirical evaluation of gated recurrent neural networks on sequence modeling. arXiv preprint arXiv:1412.3555.
[50]. Lu, R., & Duan, Z. (2017). Bidirectional GRU for sound event detection. Detection and Classification of Acoustic Scenes and Events, 1-3.
[51]. Minh, D. L., Sadeghi-Niaraki, A., Huy, H. D., Min, K., & Moon, H. (2018). Deep learning approach for short-term stock trends prediction based on two-stream gated recurrent unit network. Ieee Access, 6, 55392-55404.
[52]. Fu, R., Zhang, Z., & Li, L. (2016, November). Using LSTM and GRU neural network methods for traffic flow prediction. In 2016 31st Youth academic annual conference of Chinese association of automation (YAC) (pp. 324-328). IEEE.
[53]. Ramachandran, P., Zoph, B., & Le, Q. V. (2017). Searching for activation functions. arXiv preprint arXiv:1710.05941.
[54]. Han, J., & Moraga, C. (1995, June). The influence of the sigmoid function parameters on the speed of backpropagation learning. In International workshop on artificial neural networks (pp. 195-201). Berlin, Heidelberg: Springer Berlin Heidelberg.
[55]. Fan, E. (2000). Extended tanh-function method and its applications to nonlinear equations. Physics Letters A, 277(4-5), 212-218.
[56]. Agarap, A. F. (2018). Deep learning using rectified linear units (relu). arXiv preprint arXiv:1803.08375.
[57]. Sudre, C. H., Li, W., Vercauteren, T., Ourselin, S., & Jorge Cardoso, M. (2017). Generalised dice overlap as a deep learning loss function for highly unbalanced segmentations. In Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support: Third International Workshop, DLMIA 2017, and 7th International Workshop, ML-CDS 2017, Held in Conjunction with MICCAI 2017, Québec City, QC, Canada, September 14, Proceedings 3 (pp. 240-248). Springer International Publishing.
[58]. Christoffersen, P., & Jacobs, K. (2004). The importance of the loss function in option valuation. Journal of Financial Economics, 72(2), 291-318.
[59]. Mao, A., Mohri, M., & Zhong, Y. (2023, July). Cross-entropy loss functions: Theoretical analysis and applications. In International Conference on Machine Learning (pp. 23803-23828). PMLR.
[60]. Hodson, T. O. (2022). Root mean square error (RMSE) or mean absolute error (MAE): When to use them or not. Geoscientific Model Development Discussions, 2022, 1-10.
[61]. Van Erven, T., & Harremos, P. (2014). Rényi divergence and Kullback-Leibler divergence. IEEE Transactions on Information Theory, 60(7), 3797-3820.
[62]. Gal, Y., & Ghahramani, Z. (2016, June). Dropout as a bayesian approximation: Representing model uncertainty in deep learning. In international conference on machine learning (pp. 1050-1059). PMLR.
[63]. Sowell, F. (1992). Modeling long-run behavior with the fractional ARIMA model. Journal of monetary economics, 29(2), 277-302.
[64]. Wagenmakers, E. J., & Farrell, S. (2004). AIC model selection using Akaike weights. Psychonomic bulletin & review, 11, 192-196.
[65]. Vrieze, S. I. (2012). Model selection and psychological theory: a discussion of the differences between the Akaike information criterion (AIC) and the Bayesian information criterion (BIC). Psychological methods, 17(2), 228.
[66]. Ariyo, A. A., Adewumi, A. O., & Ayo, C. K. (2014, March). Stock price prediction using the ARIMA model. In 2014 UKSim-AMSS 16th international conference on computer modelling and simulation (pp. 106-112). IEEE.
[67]. He, Z., & Tao, H. (2018). Epidemiology and ARIMA model of positive-rate of influenza viruses among children in Wuhan, China: A nine-year retrospective study. International Journal of Infectious Diseases, 74, 61-70.
[68]. Somvanshi, V. K., Pandey, O. P., Agrawal, P. K., Kalanker, N. V., Prakash, M. R., & Chand, R. (2006). Modeling and prediction of rainfall using artificial neural network and ARIMA techniques. J. Ind. Geophys. Union, 10(2), 141-151.
[69]. Bollerslev, T., Engle, R. F., & Nelson, D. B. (1994). ARCH models. Handbook of econometrics, 4, 2959-3038.
[70]. Bauwens, L., Laurent, S., & Rombouts, J. V. (2006). Multivariate GARCH models: a survey. Journal of applied econometrics, 21(1), 79-109.
[71]. Franses, P. H., & Van Dijk, D. (1996). Forecasting stock market volatility using (non‐linear) Garch models. Journal of forecasting, 15(3), 229-235.
[72]. Alexander, C., & Lazar, E. (2006). Normal mixture GARCH (1, 1): Applications to exchange rate modelling. Journal of Applied Econometrics, 21(3), 307-336.
[73]. Mohammadi, H., & Su, L. (2010). International evidence on crude oil price dynamics: Applications of ARIMA-GARCH models. Energy Economics, 32(5), 1001-1008.
[74]. So, M. K., & Philip, L. H. (2006). Empirical analysis of GARCH models in value at risk estimation. Journal of International Financial Markets, Institutions and Money, 16(2), 180-197.
[75]. Bright, J. M. (2019). Solcast: Validation of a satellite-derived solar irradiance dataset. Solar energy, 189, 435-449.
[76]. Patro, S. G. O. P. A. L., & Sahu, K. K. (2015). Normalization: A preprocessing stage. arXiv preprint arXiv:1503.06462.
[77]. Lin, K. C., Hsu, J. Y., Wang, H. W., & Chen, M. Y. (2024). Early fault prediction for wind turbines based on deep learning. Sustainable Energy Technologies and Assessments, 64, 103684.
[78]. Li, D., Jiang, F., Chen, M., & Qian, T. (2022). Multi-step-ahead wind speed forecasting based on a hybrid decomposition method and temporal convolutional networks. Energy, 238, 121981.
[79]. Cao, J., Li, Z., & Li, J. (2019). Financial time series forecasting model based on CEEMDAN and LSTM. Physica A: Statistical mechanics and its applications, 519, 127-139.