Research Article
BibTex RIS Cite

RASSAL ORMAN REGRESYONU VE DESTEK VEKTÖR REGRESYONU İLE PİYASA TAKAS FİYATININ TAHMİNİ

Year 2021, Volume: 3 Issue: 1, 1 - 15, 30.06.2021
https://doi.org/10.51541/nicel.832164

Abstract

Antik çağdan beri varlığı bilinen statik elektrik 1880’de New York’ta üretilen elektrik ile farklı bir anlam kazanarak insan hayatının vazgeçilmez bir unsuru olmuştur. Günümüzde, temel ihtiyaç alanına girmiş olan elektriğin üretiminden dağıtımına kadar önceleri devlet tekeliyle gerçekleştirilse de özellikle 1980’lı yıllardan itibaren elektrik piyasası serbestleştirilmeye başlanarak rekabetçi bir yapıya dönüşmesi amaçlanmıştır. Serbestleşme adımları başta Şili olmak üzere, İngiltere, Avustralya, Yeni Zelanda ve Baltık ülkelerinde gerçekleşmiş ve günümüzde de bu dönüşüm süreci devam etmektedir. Ülkemizde ise elektrik piyasasındaki serbestleşme çalışmaları tam olarak 2000’li yıllarda gerçekleşmeye başlamıştır. 2015 yılında EPDK’dan aldığı piyasa işletim lisansı ile Enerji Piyasaları İşletme Anonim Şirketi (EPİAŞ) faaliyete geçerek elektrik piyasasının serbestleştirilmesinde önemli bir adım atılmıştır. Bu çalışmada, EPİAŞ tarafından işletilmekte olan Gün Öncesi Piyasası’nda belirlenen saatlik Piyasa Takas Fiyatının (PTF) tahmin edilmesi amaçlanmıştır. PTF’nin geçmiş değerlerinin ve gün öncesi piyasasında oluşan işlem hacminin PTF tahminindeki başarısı araştırılmıştır. Tahmin yöntemi olarak, makine öğrenmesi yöntemlerinden rassal orman regresyonu ve destek vektör regresyonu kullanılmıştır. Analiz sonucunda, makine öğrenmesi yöntemlerinin tahmin performanslarının karşılaştırılmasında literatürde sıklıkla kullanılan RMSE, MAE ve MAPE kriterlerine göre rassal orman regresyon yöntemi ile gerçekleştirilen ve işlem hacminin de dahil olduğu değişken grubu PTF’yi en iyi tahmin eden model (RFR-2.grup) olmuştur. Bu çalışma ile işlem hacminin PTF için önemli bir değişken olduğu belirlenmiş olup PTF tahmin çalışmalarında diğer yöntemlere göre görece daha az kullanılan rassal orman regresyonunda bu yöntemler kadar önemli olduğu görülmüştür.

References

  • Breiman, L. (2001), Random forests. Machine learning, 45(1), 5-32.
  • Catalão, J., Mariano, S., Mendes, V. and Ferreira, L. (2005), An artificial neural network approach for day-ahead electricity prices forecasting, WSEAS Transactions on Systems, 4(4), 451-454.
  • Conejo, A. J., Plazas, M. A., Espinola, R. and Molina, A. B. (2005), Day-ahead electricity price forecasting using the wavelet transform and ARIMA models, IEEE transactions on power systems, 20(2), 1035-1042.
  • Cutler, A., Cutler, D. R. and Stevens, J. R. (2012), Random forests, Ensemble machine learning Methods and Applications, Springer, Boston, MA.
  • Dangeti, P. (2017), Statistics for machine learning, Packt Publishing Ltd., Birmingham, UK.
  • Davò, F., Vespucci, M. T., Gelmini, A., Grisi, P. and Ronzio, D. (2016, October). Forecasting Italian electricity market prices using a Neural Network and a Support Vector Regression. In 2016 AEIT International Annual Conference (AEIT), 1-6.
  • de Marcos, R. A., Bello, A. and Reneses, J. (2019), Electricity price forecasting in the short term hybridising fundamental and econometric modelling, Electric Power Systems Research, 167, 240-251.
  • Ding, L. and Ge, Q. (2018), Electricity market clearing price forecast based on adaptive Kalman filter. In 2018 International Conference on Control, Automation and Information Sciences (ICCAIS), 417-421.
  • Elektrik Piyasası Dengeleme ve Uzlaştırma Yönetmeliği, https://www.epias.com.tr/mevzuat/dengeleme-ve-uzlastirma-yonetmeligi/, Erişim Tarihi:11.10.2020.
  • EPİAŞ Şeffaflık Platformu, https://seffaflik.epias.com.tr/transparency/piyasalar/gop/ptf.xhtml, Erişim Tarihi:11.10.2020.
  • EPİAŞ, https://www.epias.com.tr/gun-oncesi-piyasasi/surecler/, Erişim Tarihi:11.10.2020.
  • GÖP kullanıcı kılavuzu, https://www.epias.com.tr/gun-oncesi piyasasi/gop-kullanici-kilavuzu/, Erişim Tarihi:12.10.2020.
  • Hastie, T., Tibshirani, R. and Friedman, J. (2009). The elements of statistical learning: data mining, inference, and prediction. Springer Science & Business Media.
  • Kumar, N. (2016). Market clearing price prediction using ANN in indian electricity markets. In 2016 International Conference on Energy Efficient Technologies for Sustainability (ICEETS), 454-458.
  • McGlynn, D., Coleman, S., Kerr, D. and McHugh, C. (2018), Day-Ahead Price Forecasting in Great Britain’s BETTA Electricity Market. In 2018 IEEE Symposium Series on Computational Intelligence (SSCI), 2112-2116.
  • Mohandes, M. (2002), Support vector machines for short‐term electrical load forecasting. International Journal of Energy Research, 26(4), 335-345.
  • Nargale, K. K. and Patil, S. B. (2016), Day ahead price forecasting in deregulated electricity market using Artificial Neural Network, In 2016 International Conference on Energy Efficient Technologies for Sustainability (ICEETS), 527-532.
  • Sahay, K. B. and Tripathi, M. M. (2014), Day ahead hourly load forecast of PJM electricity market and ISO New England market by using artificial neural network, In ISGT 2014, 1-5.
  • Saini, D., Saxena, A.,and Bansal, R. C. (2016). Electricity price forecasting by linear regression and SVM, In 2016 International Conference on Recent Advances and Innovations in Engineering (ICRAIE), 1-7.
  • Santamaría-Bonfil, G., Frausto-Solís, J. and Vázquez-Rodarte, I. (2015), Volatility forecasting using support vector regression and a hybrid genetic algorithm, Computational Economics, 45(1), 111-133.
  • Smola, A. J. and Schölkopf, B. (2004), A tutorial on support vector regression. Statistics and computing, 14(3), 199-222.
  • Sun, W. and Zhang, J. (2008), Forecasting day ahead spot electricity prices based on GASVM, In 2008 International Conference on Internet Computing in Science and Engineering, 73-78.
  • Tang, Q. and Gu, D. (2009), Day-ahead electricity prices forecasting using artificial neural networks, In 2009 International Conference on Artificial Intelligence and Computational Intelligence, 2, 511-514.
  • Tat, A. N. (2018), Electricity Price Forecasting Using Monte Carlo Simulation: The Case of Lithuania, Ekonomika (Economics), 97(1), 76-86.
  • Tay, F. E. and Cao, L. (2001), Application of support vector machines in financial time series forecasting, Omega, 29(4), 309-317.
  • Yan, X. and Chowdhury, N. A. (2013), Mid-term electricity market clearing price forecasting: A hybrid LSSVM and ARMAX approach. International Journal of Electrical Power & Energy Systems, 53, 20-26.

FORECASTING OF MARKET CLEARING PRICE WITH RANDOM FOREST REGRESSION AND SUPPORT VECTOR REGRESSION

Year 2021, Volume: 3 Issue: 1, 1 - 15, 30.06.2021
https://doi.org/10.51541/nicel.832164

Abstract

Electricity known as static electricity until nineteenth century took on a new meaning with it was generated in New York in 1880. Electricity have been an indispensable instrument for human life. Initially, governments undertook the electricity process from generating to distribution. However, in 1980’s, also in Turkey at the beginning of the 2000’s, countries such as Chile, England, Australia started to liberalize their electricity markets for competition. This paper aims to predict hourly Market Clearing Price (MCP) announcing in Day Ahead Market being operated by Turkish Energy Exchange (EXIST) with market operating licence dated from 2015. It is researched that lagged values of MCP and trade value of day ahead market how prediction success on MCP. As prediction method, random forest and support vector machine of machine learning methods was used. Analysis period involve 1 Jan 2019 00:00 and 10 Mar 2020 23:00 and consist of 10440 data divided into two subset as training set (%84) and test set (%16). K-fold Cross-validation method is used for describing best parameter. Analysis was implemented in R program with caret and e1071 package. As a result of the analysis, according to the RMSE, MAPE, MAE using frequently in literature for comparing forecast performance, the best method and the best variable group which predict MCP is respectively random forest regression and the group including trade value. Therefore, this paper demonstrated that trade value is important variable for MCP and random forest is important method just as other methods used prediction of MCP.

References

  • Breiman, L. (2001), Random forests. Machine learning, 45(1), 5-32.
  • Catalão, J., Mariano, S., Mendes, V. and Ferreira, L. (2005), An artificial neural network approach for day-ahead electricity prices forecasting, WSEAS Transactions on Systems, 4(4), 451-454.
  • Conejo, A. J., Plazas, M. A., Espinola, R. and Molina, A. B. (2005), Day-ahead electricity price forecasting using the wavelet transform and ARIMA models, IEEE transactions on power systems, 20(2), 1035-1042.
  • Cutler, A., Cutler, D. R. and Stevens, J. R. (2012), Random forests, Ensemble machine learning Methods and Applications, Springer, Boston, MA.
  • Dangeti, P. (2017), Statistics for machine learning, Packt Publishing Ltd., Birmingham, UK.
  • Davò, F., Vespucci, M. T., Gelmini, A., Grisi, P. and Ronzio, D. (2016, October). Forecasting Italian electricity market prices using a Neural Network and a Support Vector Regression. In 2016 AEIT International Annual Conference (AEIT), 1-6.
  • de Marcos, R. A., Bello, A. and Reneses, J. (2019), Electricity price forecasting in the short term hybridising fundamental and econometric modelling, Electric Power Systems Research, 167, 240-251.
  • Ding, L. and Ge, Q. (2018), Electricity market clearing price forecast based on adaptive Kalman filter. In 2018 International Conference on Control, Automation and Information Sciences (ICCAIS), 417-421.
  • Elektrik Piyasası Dengeleme ve Uzlaştırma Yönetmeliği, https://www.epias.com.tr/mevzuat/dengeleme-ve-uzlastirma-yonetmeligi/, Erişim Tarihi:11.10.2020.
  • EPİAŞ Şeffaflık Platformu, https://seffaflik.epias.com.tr/transparency/piyasalar/gop/ptf.xhtml, Erişim Tarihi:11.10.2020.
  • EPİAŞ, https://www.epias.com.tr/gun-oncesi-piyasasi/surecler/, Erişim Tarihi:11.10.2020.
  • GÖP kullanıcı kılavuzu, https://www.epias.com.tr/gun-oncesi piyasasi/gop-kullanici-kilavuzu/, Erişim Tarihi:12.10.2020.
  • Hastie, T., Tibshirani, R. and Friedman, J. (2009). The elements of statistical learning: data mining, inference, and prediction. Springer Science & Business Media.
  • Kumar, N. (2016). Market clearing price prediction using ANN in indian electricity markets. In 2016 International Conference on Energy Efficient Technologies for Sustainability (ICEETS), 454-458.
  • McGlynn, D., Coleman, S., Kerr, D. and McHugh, C. (2018), Day-Ahead Price Forecasting in Great Britain’s BETTA Electricity Market. In 2018 IEEE Symposium Series on Computational Intelligence (SSCI), 2112-2116.
  • Mohandes, M. (2002), Support vector machines for short‐term electrical load forecasting. International Journal of Energy Research, 26(4), 335-345.
  • Nargale, K. K. and Patil, S. B. (2016), Day ahead price forecasting in deregulated electricity market using Artificial Neural Network, In 2016 International Conference on Energy Efficient Technologies for Sustainability (ICEETS), 527-532.
  • Sahay, K. B. and Tripathi, M. M. (2014), Day ahead hourly load forecast of PJM electricity market and ISO New England market by using artificial neural network, In ISGT 2014, 1-5.
  • Saini, D., Saxena, A.,and Bansal, R. C. (2016). Electricity price forecasting by linear regression and SVM, In 2016 International Conference on Recent Advances and Innovations in Engineering (ICRAIE), 1-7.
  • Santamaría-Bonfil, G., Frausto-Solís, J. and Vázquez-Rodarte, I. (2015), Volatility forecasting using support vector regression and a hybrid genetic algorithm, Computational Economics, 45(1), 111-133.
  • Smola, A. J. and Schölkopf, B. (2004), A tutorial on support vector regression. Statistics and computing, 14(3), 199-222.
  • Sun, W. and Zhang, J. (2008), Forecasting day ahead spot electricity prices based on GASVM, In 2008 International Conference on Internet Computing in Science and Engineering, 73-78.
  • Tang, Q. and Gu, D. (2009), Day-ahead electricity prices forecasting using artificial neural networks, In 2009 International Conference on Artificial Intelligence and Computational Intelligence, 2, 511-514.
  • Tat, A. N. (2018), Electricity Price Forecasting Using Monte Carlo Simulation: The Case of Lithuania, Ekonomika (Economics), 97(1), 76-86.
  • Tay, F. E. and Cao, L. (2001), Application of support vector machines in financial time series forecasting, Omega, 29(4), 309-317.
  • Yan, X. and Chowdhury, N. A. (2013), Mid-term electricity market clearing price forecasting: A hybrid LSSVM and ARMAX approach. International Journal of Electrical Power & Energy Systems, 53, 20-26.
There are 26 citations in total.

Details

Primary Language Turkish
Subjects Statistics
Journal Section Articles
Authors

Sinan Demirezen 0000-0002-5009-6421

Meral Çetin 0000-0003-0247-7120

Publication Date June 30, 2021
Published in Issue Year 2021 Volume: 3 Issue: 1

Cite

APA Demirezen, S., & Çetin, M. (2021). RASSAL ORMAN REGRESYONU VE DESTEK VEKTÖR REGRESYONU İLE PİYASA TAKAS FİYATININ TAHMİNİ. Nicel Bilimler Dergisi, 3(1), 1-15. https://doi.org/10.51541/nicel.832164