Araştırma Makalesi
BibTex RIS Kaynak Göster

PORTEKİZ GÜN ÖNCESİ ELEKTRİK FİYATLARININ DERİN ÖĞRENME YÖNTEMİYLE TAHMİNİ: ZAMAN SERİLERİ İÇİN SİNİRSEL TEMEL GENİŞLEME ANALİZİ MODELİ

Yıl 2025, Cilt: 28 Sayı: 4, 1902 - 1915, 03.12.2025

Burak Şafak , Oğuz Kırat , Alper Çiçek

https://doi.org/10.17780/ksujes.1736690

Öz

Elektrik piyasalarında fiyat tahmini, arz-talep dengesinin sağlanması ve piyasa katılımcılarının stratejik kararlar alabilmesi açısından kritik bir rol oynamaktadır. Gün öncesi piyasalarda fiyat dalgalanmalarının doğru tahmini, enerji ticareti ve planlama süreçlerinde önemli avantajlar sunmaktadır. Bu bağlamda, derin öğrenme tabanlı modeller, fiyat tahmini açısından büyük önem taşımaktadır. Bu çalışmada, Portekiz gün öncesi elektrik piyasasında piyasa takas fiyatı tahmini için Yenilemeli Sinir Ağı (Recurrent Neural Network, RNN), Uzun-Kısa Süreli Bellek (Long Short-Term Memory - LSTM), Kapılı Tekrarlayan Birim (Gated Recurrent Unit - GRU) ve Sinirsel Taban Genişletme Analizi (Neural Basis Expansion Analysis for Time Series - N-BEATS) modelleri kullanılmıştır. Modellerin eğitimi ve test edilmesi sürecinde, Portekiz gün öncesi elektrik piyasasından alınan gerçek piyasa verileri kullanılmış ve tahminler yedi günlük fiyat serisi üzerinden değerlendirilmiştir. Model performansları, Ortalama Kare Hatası (Mean Squared Error - MSE), Kök Ortalama Kare Hatası (Root Mean Squared Error - RMSE) ve Ortalama Mutlak Hata (Mean Absolute Error - MAE) metrikleri ile ölçülmüştür. Elde edilen sonuçlar, N-BEATS modelinin en iyi performansı sergilediğini göstermektedir. Bu model, MSE: 955,16 RMSE: 28.19 ve MAE: 24.06 değerleriyle diğer modellere üstünlük sağladığından gün öncesi piyasalarda fiyat tahmini için güçlü bir alternatiftir.

Anahtar Kelimeler

: Derin öğrenme yapay zekâ ile zaman serisi tahmini zaman serileri için sinirsel temel genişleme analizi modeli Portekiz gün öncesi elektrik fiyat tahmini uzun kısa süreli bellek

Kaynakça

  • Akdere, R., Kılıç, E., & Keçecioğlu, Ö. F. (2023). Kısmi Gölgelenme Koşullarındaki FV Sistemlerin Derin Öğrenme Tabanlı Maksimum Güç Noktası Tahmini. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, 26(3), 589–603. https://doi.org/10.17780/ksujes.1195499
  • Chung, J., Gulcehre, C., Cho, K., & Bengio, Y. (2014). Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling (Versiyon 1). arXiv. https://doi.org/10.48550/ARXIV.1412.3555
  • Çiçek, A., Güzel, S., Erdinç, O., & Catalão, J. P. S. (2021). Comprehensive survey on support policies and optimal market participation of renewable energy. Içinde Electric Power Systems Research (C. 201, s. 107522). Elsevier BV. https://doi.org/10.1016/j.epsr.2021.107522
  • Demirci, E., & Karaatlı, M. (2023). Kripto Para Fiyatlarının LSTM ve GRU Modelleri ile Tahmini, Journal of Mehmet Akif Ersoy University Economics and Administrative Sciences Faculty, Volume 10, Issue 1, Pages 134-157, https://doi.org/10.30798/makuiibf.1035314
  • Ehsani, B., Pineau, P. O., & Charlin L. (2024). Price forecasting in the Ontario electricity market via TriConvGRU hybrid model: Univariate vs. multivariate frameworks. Applied Energy, Volume 359, 122649, https://doi.org/10.1016/j.apenergy.2024.122649
  • El-Azab, H. I., Swief, R. A., El-Amary, N. H., & Temraz, H. K. (2024). Machine and deep learning approaches for forecasting electricity price and energy load assessment on real datasets. Ain Shams Engineering Journal, Volume 15, Issue 4, 102613. https://doi.org/10.1016/j.asej.2023.102613
  • Elman, J. L. (1990). Finding Structure in Time. Cognitive Science, 14(2), 179-211. https://doi.org/10.1207/s15516709cog1402_1
  • Gielen, D., Boshell, F., Saygin, D., Bazilian, M. D., Wagner, N., & Gorini, R. (2019). The role of renewable energy in the global energy transformation. Içinde Energy Strategy Reviews (C. 24, ss. 38-50). Elsevier BV. https://doi.org/10.1016/j.esr.2019.01.006
  • Gökçe, M. M., & Duman, E. (2024). A Deep Learning-Based Demand Forecasting System For Planning Electricity Generation. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, 27(2), 511–522. https://doi.org/10.17780/ksujes.1399160
  • Hochreiter, S., & Schmidhuber, J. (1997). Long Short-Term Memory. Içinde Neural Computation (C. 9, Issue 8, ss. 1735-1780). MIT Press. https://doi.org/10.1162/neco.1997.9.8.1735
  • Kabeyi, M. J. B., & Olanrewaju, O. A. (2022). Sustainable Energy Transition for Renewable and Low Carbon Grid Electricity Generation and Supply. Içinde Frontiers in Energy Research (C. 9). Frontiers Media SA. https://doi.org/10.3389/fenrg.2021.743114
  • Kannan, M., Sundareswaran, K., Nayak P. S. R., & Simon S. P. (2023). A Combined DNN-NBEATS Architecture for State of Charge Estimation of Lithium-Ion Batteries in Electric Vehicles. IEEE Transactions on Vehicular Technology, Volume 72, Issue 6. https://doi.org/10.1109/TVT.2023.3242269
  • Khan, A. A. A., Ullah, Md. H., Tabassum, R., & Kabir, Md. F. (2024, April). A Transformer-BiLSTM based Hybrid Deep Learning Approach for Day-Ahead Electricity Price Forecasting. 2024 IEEE Kansas Power and Energy Conference (KPEC), IEEE. https://doi.org/10.1109/KPEC61529.2024.10676111
  • Kırat, O., & Çiçek, A. (2024b, November). Artificial Intelligence Based Optimal Operation of Green Hydrogen Refueling. 2024 32nd Telecommunications Forum (TELFOR), Belgrade, Serbia, (pp. 1-4). IEEE. https://doi.org/ 10.1109/TELFOR63250.2024.10819143
  • Kırat, O., Çiçek, A., & Yerlikaya, T. (2024a). A New Artificial Intelligence-Based System for Optimal Electricity Arbitrage of a Second-Life Battery Station in Day-Ahead Markets. Applied Sciences, 14 (21),10032. https://doi.org/10.3390/app142110032
  • Li, J., Lin, T., Du, H., Li, Q., Fu, X., & Xu, X. (2023, Temmuz). A wind power prediction model based on optimized N-BEATS network with multivariate inputs. 2023 IEEE Power & Energy Society General Meeting (PESGM), IEEE. https://doi.org/10.1109/PESGM52003.2023.10253377
  • Li, J., Zhang, C., You, P., Yin, S., Lu, Y., & Li, C. (2024, Temmuz). A Hybrid GRU-LightGBM Model for Day-Ahead Electricity Price Forecasting. 2024 3rd International Conference on Energy and Electrical Power Systems (ICEEPS), IEEE. https://doi.org/10.1109/ICEEPS62542.2024.10693001
  • OMI, Polo Español S.A. (OMIE), Day-ahed Market Prices (2025, Şubat), https://www.omie.es/en, Erişim Tarihi: Mart 2025
  • Olivares, K. G., Challu, C., Marcjasz, G., Weron, R., & Dubrawski, A. (2023). Neural basis expansion analysis with exogenous variables: Forecasting electricity prices with NBEATSx. International Journal of Forecasting, Volume 39, Issue 2, Pages 884-900. https://doi.org/ 10.1016/j.ijforecast.2022.03.001
  • Oreshkin, B. N., Carpov, D., Chapados, N., & Bengio, Y. (2019). N-BEATS: Neural basis expansion analysis for interpretable time series forecasting. International Conference on Learning Representations. https://doi.org/10.48550/ARXIV.1905.10437
  • Paterakis, N. G., Erdinç, O., & Catalão, J. P. S. (2017). An overview of Demand Response: Key-elements and international experience. Renewable and Sustainable Energy Reviews (C. 69, ss. 871-891). Elsevier BV. https://doi.org/10.1016/j.rser.2016.11.167
  • Rahman, Md. S., Reza, H., & Kim, E. (2023, June). A Hybrid Deep Neural Network Model to Forecast Day-Ahead Electricity Prices in the USA Energy Market. 2023 IEEE World AI IoT Congress (AIIoT), IEEE. https://doi.org/10.1109/AIIoT58121.2023.10174342
  • Sefer, T., & Kaya, M. (2024). Detection of Dust on Solar Panels with Deep Learning. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, 27(4), 1451–1464. https://doi.org/10.17780/ksujes.1493906
  • Şimşek, A. İ. (2024). Comparative Analysis of Machine and Deep Learning Methods in Estimating the Turkish Electricity Market Clearing Price. Fırat Üniversitesi Mühendislik Bilimleri Dergisi. https://doi.org/10.35234/fumbd.1473145
  • Xu, Y., Huang, X., Zheng, X., Zeng, Z., & Jin, T. (2024). VMD-ATT-LSTM electricity price prediction based on grey wolf optimization algorithm in electricity markets considering renewable energy. Renewable Energy, Volume 236, 121408. https://doi.org/10.1016/j.renene.2024.121408
  • Yan, W., Wang, P., Xu, R., Han, R., Chen, E., Han, Y., & Zhang, X. (2025). A novel mid- and long-term time-series forecasting framework for electricity price based on hierarchical recurrent neural networks. Journal of the Franklin Institute, 362(6), 107590. https://doi.org/10.1016/j.jfranklin.2025.107590
  • Zhang, B., Song, C., Jiang, X., & Li, Y. (2023). Electricity price forecast based on the STL-TCN-NBEATS model. Heliyon, Volume 9, Issue 1, e13029. https://doi.org/10.1016/j.heliyon.2023.e13029

DAY-AHEAD ELECTRICITY MARKET PRICES FORECASTING IN PORTUGAL USING DEEP LEARNING: NEURAL BASIS EXPANSION ANALYSIS FOR TIME SERIES MODEL

Yıl 2025, Cilt: 28 Sayı: 4, 1902 - 1915, 03.12.2025

Burak Şafak , Oğuz Kırat , Alper Çiçek

https://doi.org/10.17780/ksujes.1736690

Öz

Price forecasting in electricity markets is crucial for maintaining supply-demand balance and guiding strategic decisions. Accurate day-ahead price predictions offer significant advantages in energy trading and planning. This study applies RNN, LSTM, GRU, and N-BEATS models to forecast market clearing prices in the Portuguese day-ahead market using real market data over a seven-day price series. Model performance was evaluated using MSE, RMSE, and MAE metrics. Results show that the N-BEATS model achieved the best performance (MSE: 955.16, RMSE: 28.19, MAE: 24.06), highlighting its potential as a strong alternative for price forecasting.

Anahtar Kelimeler

Deep learning gated recurrent unit long short-term memory modified neural basis expansion analysis for time series model Portugal day-ahead electricity price forecasting.

Kaynakça

  • Akdere, R., Kılıç, E., & Keçecioğlu, Ö. F. (2023). Kısmi Gölgelenme Koşullarındaki FV Sistemlerin Derin Öğrenme Tabanlı Maksimum Güç Noktası Tahmini. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, 26(3), 589–603. https://doi.org/10.17780/ksujes.1195499
  • Chung, J., Gulcehre, C., Cho, K., & Bengio, Y. (2014). Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling (Versiyon 1). arXiv. https://doi.org/10.48550/ARXIV.1412.3555
  • Çiçek, A., Güzel, S., Erdinç, O., & Catalão, J. P. S. (2021). Comprehensive survey on support policies and optimal market participation of renewable energy. Içinde Electric Power Systems Research (C. 201, s. 107522). Elsevier BV. https://doi.org/10.1016/j.epsr.2021.107522
  • Demirci, E., & Karaatlı, M. (2023). Kripto Para Fiyatlarının LSTM ve GRU Modelleri ile Tahmini, Journal of Mehmet Akif Ersoy University Economics and Administrative Sciences Faculty, Volume 10, Issue 1, Pages 134-157, https://doi.org/10.30798/makuiibf.1035314
  • Ehsani, B., Pineau, P. O., & Charlin L. (2024). Price forecasting in the Ontario electricity market via TriConvGRU hybrid model: Univariate vs. multivariate frameworks. Applied Energy, Volume 359, 122649, https://doi.org/10.1016/j.apenergy.2024.122649
  • El-Azab, H. I., Swief, R. A., El-Amary, N. H., & Temraz, H. K. (2024). Machine and deep learning approaches for forecasting electricity price and energy load assessment on real datasets. Ain Shams Engineering Journal, Volume 15, Issue 4, 102613. https://doi.org/10.1016/j.asej.2023.102613
  • Elman, J. L. (1990). Finding Structure in Time. Cognitive Science, 14(2), 179-211. https://doi.org/10.1207/s15516709cog1402_1
  • Gielen, D., Boshell, F., Saygin, D., Bazilian, M. D., Wagner, N., & Gorini, R. (2019). The role of renewable energy in the global energy transformation. Içinde Energy Strategy Reviews (C. 24, ss. 38-50). Elsevier BV. https://doi.org/10.1016/j.esr.2019.01.006
  • Gökçe, M. M., & Duman, E. (2024). A Deep Learning-Based Demand Forecasting System For Planning Electricity Generation. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, 27(2), 511–522. https://doi.org/10.17780/ksujes.1399160
  • Hochreiter, S., & Schmidhuber, J. (1997). Long Short-Term Memory. Içinde Neural Computation (C. 9, Issue 8, ss. 1735-1780). MIT Press. https://doi.org/10.1162/neco.1997.9.8.1735
  • Kabeyi, M. J. B., & Olanrewaju, O. A. (2022). Sustainable Energy Transition for Renewable and Low Carbon Grid Electricity Generation and Supply. Içinde Frontiers in Energy Research (C. 9). Frontiers Media SA. https://doi.org/10.3389/fenrg.2021.743114
  • Kannan, M., Sundareswaran, K., Nayak P. S. R., & Simon S. P. (2023). A Combined DNN-NBEATS Architecture for State of Charge Estimation of Lithium-Ion Batteries in Electric Vehicles. IEEE Transactions on Vehicular Technology, Volume 72, Issue 6. https://doi.org/10.1109/TVT.2023.3242269
  • Khan, A. A. A., Ullah, Md. H., Tabassum, R., & Kabir, Md. F. (2024, April). A Transformer-BiLSTM based Hybrid Deep Learning Approach for Day-Ahead Electricity Price Forecasting. 2024 IEEE Kansas Power and Energy Conference (KPEC), IEEE. https://doi.org/10.1109/KPEC61529.2024.10676111
  • Kırat, O., & Çiçek, A. (2024b, November). Artificial Intelligence Based Optimal Operation of Green Hydrogen Refueling. 2024 32nd Telecommunications Forum (TELFOR), Belgrade, Serbia, (pp. 1-4). IEEE. https://doi.org/ 10.1109/TELFOR63250.2024.10819143
  • Kırat, O., Çiçek, A., & Yerlikaya, T. (2024a). A New Artificial Intelligence-Based System for Optimal Electricity Arbitrage of a Second-Life Battery Station in Day-Ahead Markets. Applied Sciences, 14 (21),10032. https://doi.org/10.3390/app142110032
  • Li, J., Lin, T., Du, H., Li, Q., Fu, X., & Xu, X. (2023, Temmuz). A wind power prediction model based on optimized N-BEATS network with multivariate inputs. 2023 IEEE Power & Energy Society General Meeting (PESGM), IEEE. https://doi.org/10.1109/PESGM52003.2023.10253377
  • Li, J., Zhang, C., You, P., Yin, S., Lu, Y., & Li, C. (2024, Temmuz). A Hybrid GRU-LightGBM Model for Day-Ahead Electricity Price Forecasting. 2024 3rd International Conference on Energy and Electrical Power Systems (ICEEPS), IEEE. https://doi.org/10.1109/ICEEPS62542.2024.10693001
  • OMI, Polo Español S.A. (OMIE), Day-ahed Market Prices (2025, Şubat), https://www.omie.es/en, Erişim Tarihi: Mart 2025
  • Olivares, K. G., Challu, C., Marcjasz, G., Weron, R., & Dubrawski, A. (2023). Neural basis expansion analysis with exogenous variables: Forecasting electricity prices with NBEATSx. International Journal of Forecasting, Volume 39, Issue 2, Pages 884-900. https://doi.org/ 10.1016/j.ijforecast.2022.03.001
  • Oreshkin, B. N., Carpov, D., Chapados, N., & Bengio, Y. (2019). N-BEATS: Neural basis expansion analysis for interpretable time series forecasting. International Conference on Learning Representations. https://doi.org/10.48550/ARXIV.1905.10437
  • Paterakis, N. G., Erdinç, O., & Catalão, J. P. S. (2017). An overview of Demand Response: Key-elements and international experience. Renewable and Sustainable Energy Reviews (C. 69, ss. 871-891). Elsevier BV. https://doi.org/10.1016/j.rser.2016.11.167
  • Rahman, Md. S., Reza, H., & Kim, E. (2023, June). A Hybrid Deep Neural Network Model to Forecast Day-Ahead Electricity Prices in the USA Energy Market. 2023 IEEE World AI IoT Congress (AIIoT), IEEE. https://doi.org/10.1109/AIIoT58121.2023.10174342
  • Sefer, T., & Kaya, M. (2024). Detection of Dust on Solar Panels with Deep Learning. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, 27(4), 1451–1464. https://doi.org/10.17780/ksujes.1493906
  • Şimşek, A. İ. (2024). Comparative Analysis of Machine and Deep Learning Methods in Estimating the Turkish Electricity Market Clearing Price. Fırat Üniversitesi Mühendislik Bilimleri Dergisi. https://doi.org/10.35234/fumbd.1473145
  • Xu, Y., Huang, X., Zheng, X., Zeng, Z., & Jin, T. (2024). VMD-ATT-LSTM electricity price prediction based on grey wolf optimization algorithm in electricity markets considering renewable energy. Renewable Energy, Volume 236, 121408. https://doi.org/10.1016/j.renene.2024.121408
  • Yan, W., Wang, P., Xu, R., Han, R., Chen, E., Han, Y., & Zhang, X. (2025). A novel mid- and long-term time-series forecasting framework for electricity price based on hierarchical recurrent neural networks. Journal of the Franklin Institute, 362(6), 107590. https://doi.org/10.1016/j.jfranklin.2025.107590
  • Zhang, B., Song, C., Jiang, X., & Li, Y. (2023). Electricity price forecast based on the STL-TCN-NBEATS model. Heliyon, Volume 9, Issue 1, e13029. https://doi.org/10.1016/j.heliyon.2023.e13029
Toplam 27 adet kaynakça vardır.

Ayrıntılar

Birincil Dil Türkçe
Konular Derin Öğrenme, Elektrik Tesisleri
Bölüm Araştırma Makalesi
Yazarlar

Burak Şafak 0009-0001-2290-1729

Oğuz Kırat 0000-0003-2687-9351

Alper Çiçek 0000-0003-4540-2276

Gönderilme Tarihi 7 Temmuz 2025
Kabul Tarihi 7 Kasım 2025
Yayımlanma Tarihi 3 Aralık 2025
Yayımlandığı Sayı Yıl 2025 Cilt: 28 Sayı: 4

Kaynak Göster

APA Şafak, B., Kırat, O., & Çiçek, A. (2025). PORTEKİZ GÜN ÖNCESİ ELEKTRİK FİYATLARININ DERİN ÖĞRENME YÖNTEMİYLE TAHMİNİ: ZAMAN SERİLERİ İÇİN SİNİRSEL TEMEL GENİŞLEME ANALİZİ MODELİ. Kahramanmaraş Sütçü İmam Üniversitesi Mühendislik Bilimleri Dergisi, 28(4), 1902-1915. https://doi.org/10.17780/ksujes.1736690