Dalgalı Elektrik Piyasalarında Fiyat Belirleyicileri Analizi: Açıklanabilir Yapay Zeka Çerçevesi

Abstract

Yenilenebilir enerji kaynaklarının entegrasyonu, elektrik piyasalarına önemli ölçüde oynaklık (dalgalılık, volatilite) getirmiş ve geleneksel tahmin modellerini yetersiz kılmıştır. Gelişmiş makine öğrenmesi modelleri üstün doğruluk sunsa da bu modellerin kara kutu doğası, risk yönetimi için yorumlanabilir analizlere ihtiyaç duyan piyasa katılımcıları açısından kritik bir güven boşluğu yaratmaktadır. Bu çalışma, gelişmekte olan ekonomilerde Gün Öncesi Piyasası (GÖP) fiyatlarını tahmin etmek için yorumlanabilir ve açıklanabilir bir yapay zeka çerçevesi geliştirerek bu sorunu ele almaktadır. Yüksek döviz kuru oynaklığına sahip piyasalar için temsil niteliğinde bir vaka çalışması olarak Türkiye'nin kullanıldığı bu araştırmada, yedi tahmin algoritması kıyaslanmıştır. Diebold-Mariano testi aracılığıyla istatistiksel olarak anlamlı bir üstünlük sergileyen LightGBM, en uygun model olarak öne çıkmaktadır. Özellikle, modelin ekonomik teoriye uygunluğunu doğrulamak amacıyla SHAP yöntemi kullanılmış; böylece modelin doğrusal olmayan Liyakat Sırası Etkisini (Merit Order Effect) bağımsız olarak öğrenebilme yeteneği teyit edilmiştir. Şeffaf olmayan tahminleri açıklanabilir ekonomik sinyallere dönüştüren bu çerçeve, tahmin doğruluğu ile paydaş güveni arasındaki boşluğu doldurmakta ve yüksek riskli enerji ticareti ortamlarında karar verme süreçleri için güçlü bir araç sunmaktadır.

Keywords

• Elektrik Fiyat Tahmini
• Açıklanabilir Yapay Zeka
• Liyakat Sırası Etkisi
• LightGBM
• SHAP
• Enerji Ekonomisi

Supporting Institution

Türkiye Bilimsel ve Teknolojik Araştırma Kurumu (TUBITAK)

Project Number

222M440

Ethical Statement

Yazarlar herhangi bir çıkar çatışması olmadığını beyan ederler.

Thanks

Bu çalışma, Türkiye Bilimsel ve Teknolojik Araştırma Kurumu (TUBITAK) ARDEB 3501 programı kapsamında 222M440 numaralı hibe sözleşmesiyle finanse edilmiştir.

Unveiling Price Drivers in Volatile Electricity Markets: An Explainable AI Framework

Abstract

The integration of renewable energy sources has introduced significant volatility into electricity markets, rendering traditional forecasting models insufficient. While advanced Machine Learning models offer superior accuracy, their black-box nature creates a critical trust gap for market participants who require interpretable insights for risk management. This study addresses this challenge by developing an interpretable and explainable AI framework for forecasting day-ahead market prices in emerging economies. Using Türkiye as a representative case study for markets with high currency volatility, we benchmark seven forecasting algorithms. LightGBM emerges as the best model, demonstrating statistically significant superiority via the Diebold-Mariano test. Notably, we utilize SHapley Additive exPlanations (SHAP) to validate the model against economic theory, thereby confirming its ability to learn the non-linear merit order effect independently. By transforming opaque predictions into explainable economic signals, this framework bridges the gap between predictive accuracy and stakeholder trust, offering a robust tool for decision-making in high-stakes energy trading environments.

Keywords

• Electricity Price Forecasting
• Explainable Artificial Intelligence
• Merit Order Effect
• LightGBM
• SHAP
• Energy Economics

Supporting Institution

the Scientific and Technological Research Council of Türkiye (TUBITAK)

Project Number

222M440

Ethical Statement

The authors declare no conflict of interest.

Thanks

This work has received funding from the Scientific and Technological Research Council of Türkiye (TUBITAK) ARDEB 3501 program under grant agreement 222M440.

References

1. Akay MF., Abut F., Bedir B., Erdoğan M., Kanar E. Forecasting the system marginal price using long short-term memory and support vector machine. 6th International Mediterranean Science and Engineering Congress (IMSEC 2021), 25-27 October 2021, Sayfa no: 628-33, Alanya, Antalya, Türkiye.
2. Arifoğlu A., Kandemir T. Electricity price forecasting in Turkish day-ahead market via deep learning techniques. Mehmet Akif Ersoy Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi 2022; 9(2): 1433-58.
3. Arslan B., Ertuğrul İ. Çoklu regresyon, ARIMA ve yapay sinir ağı yöntemleri ile Türkiye elektrik piyasasinda fiyat tahmin ve analizi. Journal of Management Economics Research 2022; 20(1): 331-53.
4. Avci E., Ketter W., Van HE. Managing electricity price modeling risk via ensemble forecasting. The Case of Turkey Energy Policy 2018; 123(390-403).
5. Backe S., Riemer-Sørensen S., Bordvik DA., Tiwari S., Andresen C. Predictions of prices and volumes in the nordic balancing markets for electricity. 19th International Conference on the European Energy Market (EEM), 6–8 June 2023, Sayfa no: 1–6, Lappeenranta, Finland.
6. Brownlee J. How to develop a light gradient boosted machine (lightgbm) ensemble. https://machinelearningmastery.com/light-gradient-boosted-machine-lightgbm-ensemble/
7. Brownlee J. XGBoost with python: Gradient boosted trees with XGBoost and scikit-learn. San Juan, Puerto Rico: Machine Learning Mastery; 2016.
8. Chang Z., Zhang Y., Chen W. Electricity price prediction based on hybrid model of adam optimized lstm neural network and wavelet transform. Energy 2019; 187: Article no. 115804.

9. Che J., Wang J. Short-term electricity prices forecasting based on support vector regression and auto-regressive integrated moving average modeling. Energy Conversion and Management 2010; 51(10): 1911-17.
10. Conejo A., Plazas MA., Espínola R., Molina A. Day-ahead electricity price forecasting using the wavelet transform and ARIMA models. IEEE Transactions on Power Systems 2005; 20(2): 1035-1042.
11. Contreras J., Espínola R., Nogales F., Conejo A. ARIMA models to predict next-day electricity prices. IEEE Transactions on Power Systems 2003; 18(3): 1014-1020.
12. Demirezen S., Çetin M. Rassal orman regresyonu ve destek vektör regresyonu ile piyasa takas fiyatının tahmini. Nicel Bilimler Dergisi 2021; 3(1): 1-15.
13. Díaz J., Romero Á., Dorronsoro JR. Day-ahead price forecasting for the Spanish electricity market. International Journal of Interactive Multimedia and Artificial Intelligence 2019.
14. Ertaylan A., Aktaş Ö., Doğan Y. Yapay sinir ağları ile piyasa takas fiyatı tahminlemesi. Dokuz Eylül Üniversitesi Mühendislik Fakültesi Fen ve Mühendislik Dergisi 2021; 23(67): 93-105.
15. Fulcher BD., Little MA., Jones NS. Highly comparative time-series analysis: the empirical structure of time series and their methods. Journal of the Royal Society Interface 2013; 10(83): Article no. 20130048.
16. Grossi L., Nan F. Robust forecasting of electricity prices: Simulations, models and the impact of renewable sources. Technological Forecasting and Social Change 2019; 141: 305-318.
17. Gürtler M., Paulsen T. Forecasting performance of time series models on electricity spot markets: A quasi-meta-analysis. International Journal of Energy Sector Management 2018; 12(1): 103-129.
18. Haghi HV., Tafreshi SMM. Modeling and forecasting of energy prices using non-stationary Markov models versus stationary hybrid models including a survey of all methods. IEEE Canada Electrical Power Conference (EPC 2007), 25–26 October 2007, Sayfa no: 429–434, Montreal, Quebec, Canada.
19. Hayfavi A., Talasli I. Stochastic multifactor modeling of spot electricity prices. Journal of Computational and Applied Mathematics 2014; 259: 434-442.
20. Karatekin C., Başaran T. Gün öncesi piyasasında elektrik enerjisi fiyatının veri analizi ile tahmin edilmesi. Journal of the Institute of Science and Technology 2022; 12(4): 2075-2084.
21. Kölmek MA., Navruz İ. 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 2015; 23(3): 841-852.
22. Lago J., De Ridder F., De Schutter B. Forecasting spot electricity prices: Deep learning approaches and empirical comparison of traditional algorithms. Applied Energy 2018; 221: 386-405.
23. Lehna M., Scheller F., Herwartz H. Forecasting day-ahead electricity prices: A comparison of time series and neural network models taking external regressors into account. Energy Economics 2022; 106: Article no. 105742.
24. Liu H., Shen X., Tang X., Liu J. Day-ahead electricity price probabilistic forecasting based on SHAP feature selection and LSTNet quantile regression. Energies 2023; 16(13): Article no. 5152.
25. Ludwig N., Feuerriegel S., Neumann D. Putting big data analytics to work: Feature selection for forecasting electricity prices using the lasso and random forests. Journal of Decision Systems 2015; 24(1): 19-36.
26. Marcjasz G., Lago J., Weron R. Neural networks in day-ahead electricity price forecasting: single vs. multiple outputs. arXiv preprint arXiv:2008.08006; 2020.
27. Mei J., He D., Harley R., Habetler T., Qu G. A random forest method for real-time price forecasting in New York electricity market. IEEE PES General Meeting | Conference & Exposition, 27–31 July 2014, Sayfa no: 1–5, National Harbor, MD, USA.
28. Misiorek A., Trueck S., Weron R. Point and interval forecasting of spot electricity prices: Linear vs. non-linear time series models. 2006; 10(3): 1-36.
29. Muğaloğlu E., Kılıç E., Keskin H., Selçuklu SB. Dynamics of electricity price volatility and its impacts on energy investments. Renewable Energy 2025; 256: Article no. 124383.
30. Niimura T. Forecasting techniques for deregulated electricity market prices - extended survey. IEEE PES Power Systems Conference and Exposition, 29 October–1 November 2006, Sayfa no: 51–56, Atlanta, GA, USA.
31. Nogales F., Contreras J., Conejo A., Espínola R. Forecasting next-day electricity prices by time series models. IEEE Power Engineering Review 2002; 17(2): 342-348.
32. Nowotarski J., Raviv E., Trück S., Weron R. An empirical comparison of alternative schemes for combining electricity spot price forecasts. Energy Economics 2014; 46: 395-412.
33. Ozozen A., Kayakutlu G., Ketterer M., Kayalica O. A combined seasonal ARIMA and ANN model for improved results in electricity spot price forecasting: case study in Turkey. Portland International Conference on Management of Engineering and Technology (PICMET), 4–8 September 2016, Sayfa no: 2681–2690, Honolulu, HI, USA.
34. Ozyildirim C., Beyazıt M. Forecasting and modelling of electricity prices by radial basis functions: Turkish electricity market experiment. İktisat İşletme ve Finans 2014; 29(344): 31-54.
35. Peng L., Liu S., Liu R., Wang L. Effective long short-term memory with differential evolution algorithm for electricity price prediction. Energy 2018; 162: 1301-1314.
36. Plakas K., Karampinis I., Alefragis P., Birbas A., Birbas M., Papalexopoulos A. A predictive fuzzy logic model for forecasting electricity day-ahead market prices for scheduling industrial applications. Energies 2023; 16(10): Article no. 4085.
37. Prokhorenkova L., Gusev G., Vorobev A., Dorogush AV., Gulin A. Catboost: Unbiased boosting with categorical features, 32nd Conference on Neural Information Processing Systems (NeurIPS 2018), 3–8 December 2018, Sayfa no: 6639–6649, Montréal, Canada.
38. Pütz S., Kruse J., Witthaut D., Hagenmeyer V., Schäfer B. Regulatory changes in German and Austrian power systems explored with explainable artificial intelligence. 14th ACM International Conference on Future Energy Systems (e-Energy 2023), 20–23 June 2023, Sayfa no: 26–31, Orlando, FL, USA.
39. Silva AR., Fidalgo JN., Andrade JR. Easing predictors selection in electricity price forecasting with deep learning techniques. 19th International Conference on the European Energy Market (EEM), 6–8 June 2023, Sayfa no: 1–6, Lappeenranta, Finland.
40. Sirin SM., Yilmaz BN. Variable renewable energy technologies in the Turkish electricity market: Quantile regression analysis of the merit-order effect. Energy Policy 2020; 144: Article no.111660.
41. Şenocak F., Kahveci H. Periodic price averages forecasting of mcp in day-ahead Market. National Conference on Electrical, Electronics and Biomedical Engineering (ELECO), 1–3 December 2016, Sayfa no: 664–668, Bursa, Turkey.
42. Taysi ZC., Biricik G., Bozkurt ÖÖ. Evaluation of price forecast systems for Turkish electric market. 23rd Signal Processing and Communications Applications Conference (SIU), 16–19 May 2015, Sayfa no: 620–623, Malatya, Turkey.
43. Tschora L., Pierre E., Plantevit M., Robardet C. Electricity price forecasting on the day-ahead market using machine learning. Applied Energy 2022; 313: Article no. 118752.
44. Ugurlu U., Oksuz I., Tas O. Electricity price forecasting using recurrent neural networks. Energies 2018; 11(5): Article no. 1255.
45. Weron R. Electricity Price Forecasting: A review of the state-of-the-art with a look into the future. International Journal of Forecasting 2014; 30(4): 1030-1081.
46. Xie M., Sandels C., Zhu K., Nordström L. A seasonal ARIMA model with exogenous variables for elspot electricity prices in Sweden. 10th International Conference on the European Energy Market (EEM), 27–31 May 2013, Sayfa no: 1–4, Stockholm, Sweden.
47. Yıldırım MH., Özmen A., Bayrak ÖT., Weber GW. Electricity price modelling for Turkey. International Conference on Operations Research (OR 2011), 30 August–2 September 2011, Sayfa no: 39–44, Zurich, Switzerland.
48. Yu R., Gao J., Yu M., Lu W., Xu T., Zhao M., Zhang J., Zhang R., Zhang Z. Lstm-EFG for wind power forecasting based on sequential correlation features. Future Generation Computer Systems 2019; 93: 33-42.
49. Zhang J., Chuntian C. Day-ahead electricity price forecasting using artificial intelligence. IEEE Canada Electric Power Conference, 6–7 October 2008, Sayfa no: 1–5, Vancouver, BC, Canada.
50. Ziel F., Steinert R., Husmann S. Efficient modeling and forecasting of electricity spot prices. Energy Economics 2015; 47: 98-111.