Gün İçi Piyasası Elektrik Fiyat Tahmini için Eksik VerilerinTamamlanması

Türkiye regüle elektrik piyasasında, gün içi piyasasının ticaret hacmi gün geçtikçe artmaktadır. Bu durum piyasa katılımcıları içinyüksek doğrulukta tahmin yapabilmeyi önemli hale getirmektedir. Gün içi piyasasında sürekli müzayede şeklinde alışverişyapılmaktadır. Bu çalışmada, öncelikle Türkiye Gün İçi Piyasası elektrik fiyatlarının saatlik ağırlıklı ortalamaları alınarak, veri tahminproblemine hazır hale getirilmiştir. Piyasada işlem yapılmayan saatler bulunduğundan, eksik veri problemi ile karşılaşılmıştır. Elektrikfiyat tahmin literatüründe bu problemin çözümüne yönelik bir çalışma bulunmamaktadır. Bu çalışmada fiyat tahmini için kullanılacakolan eğitim verilerindeki eksiklerin nasıl tamamlanacağı üzerinde durulmuştur. Eksik veri tamamlama yöntemleri uygulamalarıkarşılaştırılmış, tek değişkenli Lasso yöntemi ile tahminler yapılarak, sonuçlar raporlanmıştır. Eksik verileri tamamlayarak tahminyapmanın sonuca istatistiksel olarak anlamlı şekilde katkısı olmuştur. Sonuçlarımız, eksik verileri Gün Öncesi Piyasası değerleri iletamamlamanın en başarılı yöntem olduğunu göstermiştir.

Data Imputation for Electricity Price Forecasting in the Intraday Market

Trading volume of the Turkish Intraday Electricity Market is rapidly increasing. This amplifies the significance of accurate electricityprice forecasts for the market players. The trading method in the intraday market is continuous trading. In this study, data is preparedto use in the price forecasting by taking a weighted average of the hourly prices. Missing value problem is encountered because of thehours without transaction. To the best of our knowledge, there is no existing study which deals with this problem in the electricity priceforecasting literature. In this article, we focused on imputing values for missing data to use them in the electricity price forecasting.Missing value methods are tested, forecasts are made by univariate lasso regression and the results are compared. Making forecasts byimputing data has increased the performance in a statistically significant manner. Our results showed that data imputation with day ahead prices is the best method.

PDF

___

Andrade, J. R., Filipe, J., Reis, M., & Bessa, R. J., 2017, “Probabilistic price forecasting for day-ahead and intraday markets: Beyond the statistical model”, Sustainability, Cilt 9, sayı 11, ss 1990.
Bicil, İ. M., 2015, Elektrik piyasasında fiyatlandırma ve Türkiye elektrik piyasasında fiyat tahmini, Doktora Tezi, Balıkesir Üniversitesi, Sosyal Bilimler Enstitüsü, Balıkesir.
Diebold, F. X., & Mariano, R. S., 2002, “Comparing predictive accuracy”, Journal of Business & economic statistics, Cilt 20, sayı 1, ss. 134-144.
Donders, A. R. T., Van Der Heijden, G. J., Stijnen, T., & Moons, K. G., 2006, “A gentle introduction to imputation of missing values”, Journal of clinical epidemiology, Cilt 59, sayı 10, ss.1087-1091.
EPİAŞ, Şeffaflık Platformu, https://seffaflik.epias.com.tr/transparency/, ziyaret tarihi:25 Ağustos 2020.
Frank, R. J., Davey, N., & Hunt, S. P., 2001, “Time series prediction and neural networks”, Journal of intelligent and robotic systems, Cilt 31, sayı 1-3, ss. 91-103.
Gunduz, S., Ugurlu, U., & Oksuz, I., 2020, “Transfer Learning for Electricity Price Forecasting”, arXiv preprint arXiv:2007.03762.
Kiesel, R., & Paraschiv, F., 2017, “Econometric analysis of 15- minute intraday electricity prices”, Energy Economics, Cilt 64, ss. 77-90.
Kölmek, M. A., & 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, Cilt 23, sayı 3, ss. 841-852.
Kulakov, S., & Ziel, F., 2019, “The impact of renewable energy forecasts on intraday electricity prices”, arXiv preprint arXiv:1903.09641.
Lago, J., De Ridder, F., & De Schutter, B., 2018, “Forecasting spot electricity prices: Deep learning approaches and empirical comparison of traditional algorithms” Applied Energy, Cilt 221, ss. 386-405.
Lepot, M., Aubin, J. B., & Clemens, F. H., 2017, “Interpolation in time series: An introductive overview of existing methods, their performance criteria and uncertainty assessment” Water, Cilt 9, sayı 10, ss. 796.
Marcjasz, G., Uniejewski, B., & Weron, R., 2020”, “Beating the Naïve—Combining LASSO with Naïve Intraday Electricity Price Forecasts”, Energies, Cilt 13, sayı 7, ss. 1667.
Monteiro, C., Ramirez-Rosado, I. J., Fernandez-Jimenez, L. A., & Conde, P., 2016, “Short-term price forecasting models based on artificial neural networks for intraday sessions in the Iberian electricity market” Energies, Cilt 9, sayı 9, ss. 721.
Narajewski, M., & Ziel, F., 2019, “Econometric modelling and forecasting of intraday electricity prices” Journal of Commodity Markets, Cilt 19, ss. 100107.
Norazian, M. N., Shukri, Y. A., & Azam, R. N., 2008, “Estimation of missing values in air pollution data using single imputation techniques”, ScienceAsia, Cilt 34, ss. 341-345
Nowotarski, J., & Weron, R., 2018, “Recent advances in electricity price forecasting: A review of probabilistic forecasting”, Renewable and Sustainable Energy Reviews, Cilt 81, ss. 1548-1568.
Oksuz, I., & Ugurlu, U., 2019, “Neural network based model comparison for intraday electricity price forecasting” Energies, Cilt 12, sayı 23, ss. 4557.
Özyildirim, C., & Beyazit, M. F., 2014, “Forecasting and modelling of electricity prices by radial basis functions: Turkish electricity market experiment” Iktisat Isletme ve Finans, Cilt 29, sayı 344, ss. 31-54.
Pape, C., Hagemann, S., & Weber, C., 2016, “Are fundamentals enough? Explaining price variations in the German day-ahead and intraday power market”, Energy Economics, Cilt 54, ss. 376-387.
Shahidehpour, M., Yamin, H., & Li, Z., 2003, “Elektrik Fiyat Tahmini”, Market operations in electric power systems: forecasting, scheduling, and risk management, John Wiley & Sons, Wiley, ss 57-113
Shinde, P., & Amelin, M., 2019, “A Literature Review of Intraday Electricity Markets and Prices”. In 2019 IEEE Milan PowerTech, June, ss. 1-6.
Smyl, S., 2020, “A hybrid method of exponential smoothing and recurrent neural networks for time series forecasting”, International Journal of Forecasting, Cilt 36, sayı 1, ss. 75-85.
Talasli, I., 2012, “Stochastic Modeling of Electricity Markets”, Doktora Tezi, Middle East Technical University, Financial Mathematics, Ankara.
Tibshirani, R., 1996, “Regression shrinkage and selection via the lasso” Journal of the Royal Statistical Society: Series B (Methodological), Cilt 58, sayı 1, ss. 267-288.
TOROS, H., & Aydın, D. (2018). Prediction of Short-Term Electricity Consumption by Artificial Neural Networks Using Temperature Variables. Avrupa Bilim ve Teknoloji Dergisi, (14), 393-398.
Ugurlu, U., Oksuz, I., & Tas, O., 2018, “Electricity price forecasting using recurrent neural networks. Energies”, Cilt 11, sayı 5, ss. 1255.
Ugurlu, U., Tas, O., Kaya, A., & Oksuz, I., 2018, “The financial effect of the electricity price forecasts’ inaccuracy on a hydrobased generation company, Energies”, Cilt 11, sayı 8, ss. 2093.
Uniejewski, B., Marcjasz, G., & Weron, R., 2019, “Understanding intraday electricity markets: Variable selection and very short-term price forecasting using LASSO”, International Journal of Forecasting, Cilt 35, sayı 4, ss. 1533-1547.
Weron, R., 2014, “Electricity price forecasting: A review of the state-of-the-art with a look into the future” International journal of forecasting, Cilt 30, sayı 4, ss. 1030-1081.
Yorulmus, H., Ugurlu, U., & Tas, O., 2018, “A Long Short Term Memory Application On The Turkish Intraday Electricity Price Forecasting”, PressAcademia Procedia, Cilt 7, sayı 1, ss. 126-130.
Zareipour, H., Bhattacharya, K., & Canizares, C. A., 2007 “Electricity market price volatility: The case of Ontario”, Energy policy, Cilt 35, sayı 9, ss. 4739-4748.
Ziel, F., 2016 “Forecasting electricity spot prices using lasso: On capturing the autoregressive intraday structure” IEEE Transactions on Power Systems, Cilt 31, sayı 6, ss. 4977- 4987.
Ziel, F., & Weron, R., 2018, “Day-ahead electricity price forecasting with high-dimensional structures: Univariate vs. multivariate modeling frameworks”, Energy Economics, Cilt 70, ss. 396-420.