DERİN ÖĞRENME TEKNİKLERİYLE TÜRKİYE GÜN ÖNCESİ PİYASASINDA ELEKTRİK FİYAT TAHMİNİ

Year 2022, Volume: 9 Issue: 2, 1433 - 1458, 29.07.2022

Arif Arifoğlu Tuğrul Kandemir

https://doi.org/10.30798/makuiibf.1097686

Abstract

Gün Öncesi Piyasası, elektrik piyasası katılımcılarına gerçek zamandan bir gün öncesinde ticaret yapma imkânı sunan bir piyasadır. Gün Öncesi Piyasasında her saat için ayrı bir Piyasa Takas Fiyatı oluşturulmaktadır. Bu çalışmada, saatlik Piyasa Takas Fiyatının derin öğrenme teknikleri kullanılarak tahmin edilmesi amaçlanmıştır. Bu doğrultuda MLP, CNN, LSTM ve GRU modelleri ile 24 saatlik Piyasa Takas Fiyatı tahmin edilmiştir. Elde edilen sonuçlara göre, LSTM 8,15 MAPE değeri ile en iyi ortalama tahmin performansına sahip olmuştur. LSTM’i 8,44 MAPE değeri ile MLP, 8,72 MAPE değeri ile GRU ve 9,27 MAPE değeri ile CNN takip izlemiştir. Bu çalışmada kullanılan meteorolojik değişkenler için yenilebilir kaynaklarla üretim yapan santrallerin yoğun olduğu iller seçilmiştir. Yenilenebilir kaynaklarla elektrik üretimine olan eğilimin gelecekte daha da artması beklenmektedir. Bu bağlamda, piyasa katılımcıları için elektrik fiyat tahmininde bu kaynaklarla gerçekleşen üretimi etkileyebilecek faktörlerin göz önüne alınmasının önemli olduğu düşünülmektedir.

Keywords

Gün Öncesi Piyasası, Fiyat Tahmini, Piyasa Takas Fiyatı, Derin Öğrenme

Supporting Institution

AFYON KOCATEPE ÜNİVERSİTESİ BİLİMSEL ARAŞTIRMA PROJELERİ BİRİMİ

Project Number

18.SOS.BİL.08

ELECTRICITY PRICE FORECASTING IN TURKISH DAY-AHEAD MARKET VIA DEEP LEARNING TECHNIQUES

Abstract

Day-Ahead Market offers electricity market participants the opportunity to trade electricity one day ahead of real-time. For each hour, a separate Market Clearing Price is created in Day-Ahead Market. This study aims to predict the hourly Market Clearing Price using deep learning techniques. In this context, 24-hour Market Clearing Prices were forecasted with MLP, CNN, LSTM, and GRU. LSTM had the best average forecasting performance with an 8.15 MAPE value, according to the results obtained. MLP followed the LSTM with 8.44 MAPE, GRU with 8.72 MAPE, and CNN with 9.27 MAPE. In the study, the provinces where the power plants producing with renewable resources are dense were selected for meteorological variables. It is expected that the trend towards electricity generation with renewable resources will increase in the future. In this context, it is thought important for market participants to consider the factors that may affect the production with these resources in the electricity price forecasting.

Keywords

Day-Ahead Market, Price Forecasting, Market Clearing Price, Deep Learning

Project Number

18.SOS.BİL.08

References

• Aineto, D., Sanchez, J. I., Zuniga, L. G. L., Onaindia, E. and Urcheguia, J. F. (2019). On The Influence of Renewable Energy Sources in Electricity Price Forecasting in The Iberian Market. Energies, 12(11), 1-20. https://doi.org/10.3390/en12112082
• Althelaya, K. A., El-Alfy, E. M. and Mohammed, S. (2018). Stock Market Forecast Using Multivariate Analysis With Bidirectional and Stacked (LSTM, GRU). 21. Saudi Computer Society National Computer Conference (NCC), Riyad. https://doi.org/10.1109/NCG.2018.8593076
• Amjady, N. and Hemmati, M. (2006). Energy Price Forecasting: Problems and Proposals For Such Predictions. IEEE Power Energy Magazine, March-April Issue, 20-29. https://doi.org/10.1109/MPAE.2006.1597990
• Anbazhagan, S. and Kumarappan, N. (2014). Day-Ahead Deregulated Electricity Market Price Forecasting Using Neural Network Input Featured by DCT. Energy Conversion and Management, 78, 711-719. https://doi.org/10.1016/j.enconman.2013.11.031
• Anochi, J. A. and Velho, H. F. C. (2016). Mesoscale Precipitation Climate Prediction for Brazilian South Region by Artificial Neural Networks. American Journal of Environmental Engineering, 6(4A), 94-102. https://doi.org/10.5923/s.ajee.201601.14
• Bento, P. M. R., Pombo, J. A. N., Calado, M. R. A. and Mariano, S. J. P. S. (2018). A Bat Optimized Neural Network and Wavelet Transform Approach for Shortterm Price Forecasting. Applied Energy, 210, 88-97. https://doi.org/10.1016/j.apenergy.2017.10.058
• Catalao, J. P. S., Mariano, S. J. P. S., Mendes, V. M. F. and Ferreira, L. A. F. M. (2007). Short-Term Electricity Prices Forecasting in a Competitive Market: A Neural Network Approach. Electric Power Systems Research, 77, 1297-1304. https://doi.org/10.1016/j.epsr.2006.09.022
• Cervone, A., Santini, E., Teodori, S. and Romito, D. Z. (2014). Electricity Price Forecast: A Comparison of Different Models to Evaluate the Single National Price in the Italian Energy Exchange Market. International Journal of Energy Economics and Policy, 4(4), 744-758.
• Chaabane, N. (2014). A Hybrid ARFIMA and Neural Network Model for Electricity Price Prediction. Electrical Power and Energy Systems, 55, 187-194. https://doi.org/10.1016/j.ijepes.2013.09.004
• Chang, Z., Zhang, Y. and Chen, W. (2018). Effective Adam-Optimized LSTM Neural Network for Electricity Price Forecasting. IEEE 9th International Conference on Software Engineering and Service Science, Beijing. https://doi.org/10.1109/ICSESS.2018.8663710
• Cheng, H. Y., Kuo, P. H., Shen, Y. and Huang, C. J. (2020). Deep Convolutional Neural Network Model for Short-Term Electricity Price Forecasting. Retrieved from: https://arxiv.org/ftp/arxiv/papers/2003/2003.07202.pdf
• Chung, J., Gülçehre, Ç., Cho, K. H. and Bengio, Y. (2014). Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling. Retrieved from: https://arxiv.org/pdf/1412.3555.pdf
• Contreras, J., Espinola, R., Nogales, F. J. and Conejo, A. J. (2003). ARIMA Models to Predict Next-Day Electricity Prices. IEEE Transactions Power Systems, 18(3), 1014-1020. https://doi.org/10.1109/TPWRS.2002.804943
• Cuaresma, J. C., Hlouskova, J., Kossmeier, S. and Obersteiner, M. (2004). Forecasting Electricity Spotprices Using Linear Univariate Time-Series Models. Applied Energy, 77(1), 87-106. https://doi.org/10.1016/S0306-2619(03)00096-5
• Devir, K. (2017). Türk Elektrik Piyasasının İşleyişi. Bursa, Dora Publications.
• EPİAŞ (2016a). Elektrik Piyasası Özet Bilgiler Raporu 2016. Retrieved from: https://www.epias.com.tr/wp-content/uploads/2017/03/EPIAS_2016_Yili_ Elektrik_Piyasasi_Ozet_Bilgiler_Raporu.pdf
• EPİAŞ (2016b). Gün Öncesi Elektrik Piyasası Piyasa Takas Fiyatı Belirleme Yöntemi. İstanbul, EPİAŞ Publications.
• EPİAŞ (2017). Elektrik Piyasası Özet Bilgiler Raporu 2017. Retrieved from: https://www.epias.com.tr/wp-content/uploads/2018/03/EPIAS_2017_Yillik_ Bulten_V2.pdf
• EPİAŞ (2018). Elektrik Piyasası Özet Bilgiler Raporu 2018. Retrieved from: https://www.epias.com.tr/wp-content/uploads/2019/03/Epias_2018_Yillik_ Bulten.pdf
• EPİAŞ (2019). Elektrik Piyasası Özet Bilgiler Raporu 2019. Retrieved from: https://www.epias.com.tr/wp-content/uploads/2020/01/EP%C4%B0AS_2019_ Y%C4%B1ll%C4%B1k_B%C3%BClten-2.pdf
• EPİAŞ (2020). Elektrik Piyasası Özet Bilgiler Raporu 2020. Retrieved from: https://www.epias.com.tr/wp-content/uploads/2021/01/EPIAS_2020_Yillik_ Bulten_vson.pdf
• Fu, R., Zhang, Z. and Li, L. (2016). Using LSTM and GRU Neural Network Methods for Traffic Flow Prediction. 31. Youth Academic Annual Conference of Chinese Association of Automation (YAC), Wuhan. https://doi.org/10.1109/YAC.2016.7804912
• Girish, G. P. (2016). Spot Electricity Price Forecasting in Indian Electricity Market Using Autoregressive-GARCH Models. Energy Strategy Reviews, 11(12), 52-57. https://doi.org/10.1016/j.esr.2016.06.005
• Goodfellow, I., Bengio, Y. and Courville, A. (2016). Deep Learning. London, MIT Press.
• Guo, X., Zhao, Q., Zheng, D., Ning, Y. and Gao, Y. (2020). A Short-Term Load Forecasting Model of Multi-Scale CNN-LSTM Hybrid Neural Network Considering The Real-Time Electricity Price. Energy Reports, 6, 1046-1053. https://doi.org/10.1016/j.egyr.2020.11.078
• Gupta, P. and Sinha, N. K. (2000). Neural Networks for Identification of Nonlinear Systems: An Overview. In N. K. Sinha and M. M. Gupta (Eds.), Soft Computing and Intelligent Systems: Theory and Applications (337-356). Cambridge, Academic Press.
• Gündüz, S., Uğurlu, U. and Öksüz, İ. (2020). Türkiye Gün Öncesi Elektrik Piyasasında Kodlayıcı Kod Çözücü Temelli Yinelemeli Sinir Ağlarıyla Fiyat Tahmini. 28th Signal Processing and Communications Applications Conference (SIU), Gaziantep.
• Hassim, Y. M. M. and Ghazali, R. (2012). Training a Functional Link Neural Network Using an Artificial Bee Colony for Solving a Classification Problems. Retrieved from: https://arxiv.org/ftp/arxiv/papers/1212/1212.6922.pdf
• Hong, Y. Y., Taylar, J. V. and Fajardo, A. C. (2020). Locational Marginal Price Forecasting in a Day-Ahead Power Market Using Spatiotemporal Deep Learning Network. Sustainable Energy, Grids and Networks, 24, 1-14. https://doi.org/10.1016/j.segan.2020.100406
• Huang, C. J., Shen, Y., Chen, Y. H. and Chen, H. C. (2020). A Novel Hybrid Deep Neural Network Model for Short-Term Electricity Price Forecasting. International Journal of Energy Research, 45(2), 2511-2532. https://doi.org/10.1002/er.5945
• Huisman, R., Huurman, C. and Mahieu, R. (2007). Hourly Electricity Prices in Day-Ahead Markets. Energy Economics, 29, 240-248. https://doi.org/10.1016/j.eneco.2006.08.005
• Indolia, S., Goswami, A. K., Mishra, S. P. and Asopa, P. (2018). Conceptual Understanding of Convolutional Neural Network- A Deep Learning Approach. Procedia Computer Science, 132, 679-688. https://doi.org/10.1016/j.procs.2018.05.069
• Jahangir, H., Tayarani, H., Baghali, S., Ahmadian, A. Elkamel, A., Golkar, M. A. and Castilla, M. (2020). A Novel Electricity Price Forecasting Approach Based on Dimension Reduction Strategy and Rough Artificial Neural Networks. IEEE Transactions on Industrial Informatics, 16(4), 2369-2381. https://doi.org/10.1109/TII.2019.2933009
• Keynia, F. (2012). A New Feature Selection Algorithm and Composite Neural Network for Electricity Price Forecasting. Engineering Applications of Artificial Intelligence, 25, 1687-1697. https://doi.org/10.1016/j.engappai.2011.12.001
• Khajeh, M. G., Maleki, A., Rosen, M. A. and Ahmadi, M. H. (2018). Electricity Price Forecasting Using Neural Networks With an Improved Iterative Training Algorithm. International Journal of Ambient Energy, 39(2), 147-158. https://doi.org/10.1080/01430750.2016.1269674
• Khan, A. and Sarfaraz, A. (2019). RNN-LSTM-GRU Based Language Transformation. Soft Computing, 23, 13007-13024. https://doi.org/10.1007/s00500-019-04281-z
• Kim, P. (2017). MATLAB Deep Learning With Machine Learning, Neural Networks and Artificial Intelligence. New York, Apress Publications.
• Klimberg, R. K., Sillup, G. P., Boyle, K. J. and Tavva, V. (2010). Forecasting Performance Measures – What Are Their Practıcal Meaning?. In K. D. Lawrence and R. K. Klimberg (Eds.), Advances in Busıness and Management Forecasting (137-147). Bingley, Emeralds Publications.
• Kouhi, S. and Keynia, F. (2013). A New Cascade NN Based Method to Short-Term Load Forecast in Deregulated Electricity Market. Energy Conversion and Management, 71, 76-83. https://doi.org/10.1016/j.enconman.2013.03.014
• Kölmek, M. A. and Navruz, İ. (2015). Forecasting The Day-Ahead Price in Electricity Balancing and Settlement Market of Turkey by Using Artificial Neural Networks. Turkish Journal of Electrical Engineering & Computer Sciences, 23, 841-852. https://doi.org/ 10.3906/elk-1212-136
• Kristiansen, T. (2014). A Time Series Spot Price Forecast Model for the Nord Pool Market. Electrical Power and Energy Systems, 61, 20-26. https://doi.org/10.1016/j.ijepes.2014.03.007
• Krizhevsky, A., Sutskever, I. and Hinton, G. E. (2017). ImageNet Classification With Deep Convolutional Neural Networks. Communications of The Acm, 60(6), 84-90. https://doi.org/10.1145/3065386
• Kumar, S., Hussain, L.,Banarjee, S. and Reza, M. (2018). Energy Load Forecasting Using Deep Learning Approach- LSTM and GRU in Spark Cluster. 5. International Conference on Emerging Applications of Information Technology (EAIT), Kolkata. https://doi.org/10.1109/EAIT.2018.8470406
• Kuo, P. H. and Huang, C. J. (2018). An Electricity Price Forecasting Model by Hybrid Structured Deep Neural Networks. Sustainability, 10(4), 1-17. https://doi.org/10.3390/su10041280
• Lago, J., Ridder, F. D. and Schutter, B. D. (2018). Forecasting Spot Electricity Prices: Deep Learning Approaches and Empirical Comparison of Traditional Algorithms. Applied Energy, 221, 386-405. https://doi.org/10.1016/j.apenergy.2018.02.069
• Lewis, N. D. (2017). Neural Networks for Time Series Forecasting With R: Intuitive Step by Step Blueprint for Beginners. South Carolina, CreateSpace Publications.
• Li, G., Liu, C. C., Lawarree, J, Gallanti, M. and Venturini, A. (2005). State-of-the-Art of Electricity Price Forecasting. International Symposium CIGRE/IEEE PES, New Orleans. https://doi.org/ 10.1109/CIGRE.2005.1532733
• Li, W. and Becker, D. M. (2021). Day-Ahead Electricity Prices Prediction Applying Hybrid Models of LSTM-Based Deep Learning Methods and Feature Selection Algorithms Under Consideration of Market Coupling. Retrieved from: https://arxiv.org/pdf/2101.05249.pdf
• Liu, H. and Shi, J. (2013). Applying ARMA–GARCH Approaches to Forecasting Short-Term Electricity Prices. Energy Economics, 37, 152-166. https://doi.org/10.1016/j.eneco.2013.02.006
• Lyu, Z., Wang, Y., Wang, J., Zhang, L., Shen, J. and Wang, X. (2019). Short-Term Electricity Price Forecasting G-LSTM Model and Economic Dispatch for Distribution System. The Fourth International Conference on Energy Engineering and Environmental Protection, Xiamen. https://doi.org/10.1088/1755-1315/467/1/012186
• Mirakyan, A., Renschhausen, M. M. and Koch, A. (2017). Composite Forecasting Approach, Application for Next-Day Electricity Price Forecasting. Energy Economics, 66, 228-237. https://doi.org/10.1016/j.eneco.2017.06.020
• Nogales, F. J., Contreras, J., Conejo, A. J. and Espinola, R. (2002). Forecasting Next-Day Electricity Prices by Time Series Models. IEEE Transactions on Power Systems, 17(2), 342-348. https://doi.org/10.1109/TPWRS.2002.1007902
• O’Shea, K. and Nash, R. (2015). An Introduction to Convolutional Neural Networks. Retrieved from: https://arxiv.org/pdf/1511.08458.pdf
• Özözen, A., Kayakutlu, G., Ketterer, M. and Kayalıca, Ö. (2016). A Combined Seasonal ARIMA and ANN Model for Improved Results in Electricity Spot Price Forecasting: Case Study in Turkey. PICMET'16 Conference: Technology Management for Social Innovation, Honolulu. https://doi.org/10.1109/PICMET.2016.7806831
• Pao, H. T. (2006). A Neural Network Approach to M-Daily-Ahead Electricity Price Prediction. Third International Symposium on Neural Networks, Chengdu. https://doi.org/10.1007/11760023_186
• Raviv, E., Bouwman, K. E. and Dijk, D. (2015). Forecasting Day-Ahead Electricity Prices: Utilizing Hourly Prices. Energy Economics, 50, 227-239. https://doi.org/10.1016/j.eneco.2015.05.014
• Sahay, K. B. (2015). One Hour Ahead Price Forecast of Ontario Electricity Market by Using ANN. 2015 International Conference on Energy Economics and Environment (ICEEE), Greater Noida. https://doi.org/10.1109/EnergyEconomics.2015.7235102
• Schnürch, S. and Wagner, A. (2020). Electricity Price Forecasting with Neural Networks on EPEX Order Books. Applied Mathematical Finance, 27(3), 189-206. https://doi.org/10.1080/1350486X.2020.1805337
• Ser, G. and Bati, C. T. (2019). Derin Sinir Ağları ile En İyi Modelin Belirlenmesi: Mantar Verileri Üzerine Keras Uygulaması. Yuzuncu Yıl University Journal of Agricultural Sciences, 29(3), 406-417. https://doi.org/10.29133/yyutbd.505086
• Shen, G., Tan, Q., Zhang, H., Zeng, P. and Xu, J. (2018). Deep Learning With Gated Recurrent Unit Networks for Financial Sequence Predictions. Procedia Computer Science, 131, 895-903. https://doi.org/10.1016/j.procs.2018.04.298
• Shewalkar, A. N. (2018). Comparison of RNN, LSTM and GRU on Speech Recognition Data. Master's Dissertation, North Dakota State University of Agriculture and Applied Science, North Dakota, USA. Retrieved from: https://library.ndsu.edu/ir/bitstream/handle/10365/29111/Comparison%20of%20RNN%2c%20LSTM%20and%20GRU%20on%20Speech%20Recognition%20Data.pdf?sequence=1&isAllowed=y
• Tan, Z., Zhang, J., Wang, J. and Xu, J. (2010). Day-Ahead Electricity Price Forecasting Using Wavelet Transform Combined With ARIMA and GARCH Models. Applied Energy, 87, 3606-3610. https://doi.org/10.1016/j.apenergy.2010.05.012
• Uğurlu, U., Öksüz, İ. and Taş, O. (2018). Electricity Price Forecasting Using Recurrent Neural Networks. Energies, 11(5), 1-23. https://doi.org/10.3390/en11051255
• Weron, R. (2014). Electricity Price Forecasting: A Review of The State-of-The-Art With a Look Into The Future. International Journal of Forecasting, 30, 1030-1081. https://doi.org/10.1016/j.ijforecast.2014.08.008
• Weron, R. and Misiorek, A. (2006). Short-Term Electricity Price Forecasting With Time Series Models: A Review and Evaluation. HSC Research Report, 6(1), 1-22.
• Xie, X., Xu, W. and Tan, H. (2018). The Day-Ahead Electricity Price Forecasting Based on Stacked CNN and LSTM. Retrieved from: https://link.springer.com/chapter/10.1007/978-3-030-02698-1_19
• Yamashita, R., Nishio, M., Do, R. K. G. and Togashi, K. (2018). Convolutional Neural Networks: An Overview and Application in Radiology. Insights Imaging, 9, 611-629. https://doi.org/10.1007/s13244-018-0639-9
• Yang, Z., Ce, L. and Lian, L. (2017). Electricity Price Forecasting by a Hybrid Model, Combining Wavelet Transform, ARMA and Kernel-Based Extreme Learning Machine Methods. Applied Energy, 190, 291-305. https://doi.org/10.1016/j.apenergy.2016.12.130
• Yarıcı, M. (2018). EPİAŞ Elektrik Piyasaları Eğitim Sunumları: Gün Öncesi Piyasası. İstanbul, EPİAŞ Publications.
• Zhang, R., Li, G. and Ma, Z. (2020a). A Deep Learning Based Hybrid Framework for Day-Ahead Electricity Price Forecasting. IEEE Access, 8, 143423-143436. https://doi.org/10.1109/ACCESS.2020.3014241
• Zhang, F., Fleyeh, H. and Bales, C. (2020b). A Hybrid Model Based on Bidirectional Long Short-Term Memory Neural Network and Catboost for Short-Term Electricity Spot Price Forecasting. Journal of the Operational Research Society, November Issue, 1-25. https://doi.org/10.1080/01605682.2020.1843976
• Zhou, S., Zhou, L., Mao, M., Tai, H. M. and Wan, Y. (2019). An Optimized Heterogeneous Structure LSTM Network for Electricity Price Forecasting, IEEE Access, 7, 108161-108173. https://doi.org/10.1109/ACCESS.2019.2932999
• Ziel, F., Steinert, R. and Husmann, S. (2015). Efiicient Modeling and Forecasting of Electricity Spot Prices. Energy Economics, 47, 98-111. https://doi.org/10.1016/j.eneco.2014.10.012
• 4628 Sayılı Kanun (2001). Retrieved from: https://www.mevzuat.gov.tr/MevzuatMetin/1.5.4628.pdf
• 6446 Sayılı Kanun (2013). Retrieved from: https://www.mevzuat.gov.tr/MevzuatMetin/1.5.6446.pdf
• https://seffaflik.epias.com.tr/transparency/, Retrieved on 01.04.2021.
• https://www.enerjiatlasi.com/gunes/, Retrieved on 01.04.2021.
• https://www.enerjiatlasi.com/ruzgar/, Retrieved on 01.04.2021.