Research Article
BibTex RIS Cite

TÜRKİYE ELEKTRİK TÜKETİMİNİN DEEP LEARNING BI-LSTM METODU İLE TAHMİNİ

Year 2022, Volume: 1 Issue: 1, 24 - 33, 31.12.2022

Abstract

Gelişmekte olan ülkeler arasında yer alan Türkiye’nin enerji tüketimi sürekli artış göstermektedir. Artan bu enerji ihtiyacına rağmen enerji üretme konusunda ise yetersiz bir ülkedir. Enerji kullanımında dışa bağımlı bir ülke konumunda olan Türkiye, sürdürülebilir enerji arzında problemler yaşamaktadır. Özellikle son dönemde Rusya’nın Avrupa ülkelerine enerji ihracatında kısıtlamalara gitmesi tüm dünyada enerji krizine neden olmaktadır. Bu nedenle tüm dünyada olduğu gibi Türkiye için de enerji arz güvenliği hayati bir role sahiptir. Bu bağlamda gelecek dönemlere ait enerji tüketim tahmini, üzerinde durulması gereken stratejik bir konudur. Çalışmada Türkiye’nin 2005 Ocak-2018 Kasım yılları arasındaki aylık enerji tüketim miktarları alınmış ve sürekli artan bir grafik seyreden elektrik tüketiminin bi-directional LSTM modelleri (ADAM, RmsProp, SGDM) ile 2019-2023 aralığında 5 yıllık tahmini yapılmıştır. Modellerde en yüksek performans RMSprop optimizasyonu ile elde edilmiştir. 2019-2020 yılları arasında aylık gerçekleşen elektrik enerjisi tüketimi verileri ile RMSprop optimizasyonu ile elde edilen aynı dönem için aylık elektrik tüketiminin tahmini verileri karşılaştırılmıştır. Optimizasyon sonucuna göre Türkiye elektrik tüketimi artmaya devam edecektir. Türkiye artacak bu ihtiyacı karşısında gerekli planlamaları hızlı bir şekilde yürürlüğe koymalıdır. Enerji tasarrufu için hane halklarının eğitiminin planlara dahil edilmesi uygun bir çözüm olabilir.

References

  • Ahmad AS, Hassan MY, Abdullah MP, Rahman HA, Hussin F, Abdullah H, Saidur R. 2014. A review on applications of ANN and SVM for building electrical energy consumption forecasting. Renewable and Sustainable Energy Reviews, 33:102-109. https://doi.org/10.1016/j.rser.2014.01.069
  • Anand A, Suganthi L. 2017. Forecasting of electricity demand by hybrid ANN-PSO models. International Journal of Energy Optimization and Engineering, 6(4):66-80. doi:10.4018/IJEOE.2017100105
  • EPDK. 2021. Electricity Market Sector Report. Ankara.
  • EPDK. 2020. Electricity Market Sector Report. Ankara.
  • EPDK, 2019. Electricity Market Sector Report, Ankara.
  • Fara L, Diaconu A, Craciunescu D, Fara S. 2021. Forecasting of energy production for photovoltaic systems based on ARIMA and ANN advanced models. International Journal of Photoenergy, 2021: e:6777488. https://doi.org/10.1155/2021/6777488
  • Cui Z, Ke R, Wang Y. 2017. Deep Stacked Bidirectional and Unidirectional LSTM Recurrent Neural Network for Network-wide Traffic Speed Prediction. Available from: https://arxiv.org/ftp/arxiv/papers/1801/1801.02143.pdf
  • Dong K, Dong X, Jiang Q. 2020. How renewable energy consumption lower global CO2 emissions? Evidence from countries with different income levels. The World Economy, 43:1665-1698. doi: 10.1111/twec.12898
  • Enerdata. 2022. https://yearbook.enerdata.net/total-energy/world-consumption-statistics.html
  • Erilli NA, Eğrioğlu E,Yolcu U, Aladağ HÇ, Uslu VR. 2010. Türkiye’de Enflasyonun İleri ve Geri Beslemeli Yapay Sinir Ağlarının Melez Yaklaşımı ile Öngörüsü. Doğuş Üniversitesi Dergisi, 11(1):42-55. Available from: https://dergipark.org.tr/tr/pub/doujournal/issue/66662/1042993
  • Fauset L. 1994. Fundamentals of Neural Network. Prentice Hall International, London
  • Gandelli A, Grimaccia F, Leva S, Mussetta M, Ogliari E. 2014. Hybrid model analysis and validation for PV energy production forecasting. 2014 International Joint Conference on Neural Networks (IJCNN), July 6-11, China. Available from: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6889786
  • Gref K, Srivastava RK, Koutnik J, Steunebrink BR, Schmindhuber J. 2017. LSTM-A Search Space Odysse, IEEE Transactions on Neural Networks and Learning Systems, 28(10):2222-2232. doi: 10.1109/TNNLS.2016.2582924
  • Haykin S. 1998. Neural Networks: A Comprehensive Foundation. Seconf Ed., Prentice hall, New Jersey.
  • Koukaras P, Bezas N, Gkaidatzis P, Ionnidis D, Tzovaras D, Tjortjis C. 2021. Introducing a novel approach in one-step ahead energy load forecasting. Sustainable Computing: Informatics and Systems, 32:100616. https://doi.org/10.1016/j.suscom.2021.100616
  • Li K, Su H, Chu J. 2011. Forecasting building energy consumption using neural networks and hybrid neuro-fuzzy system: A comparative study. Energy and Buildings, 43:2893-2899. doi:10.1016/j.enbuild.2011.07.010
  • Pazikadin RA, Rifai D, Ali K, Malik MZ, Abdalla A N, Faraj MA. 2020. Solar irradiance measurement instrumentation and power solar generation forecasting based on Artificial Neural Networks (ANN): A review of five years research trend. Science of the Total Environment, 715:e:136848. https://doi.org/10.1016/j.scitotenv.2020.136848
  • TEİAŞ. 2022. Available from: https://www.teias.gov.tr/turkiye-elektrik-uretim-iletim-istatistikleri
  • Zhang G, Patuwo BE, Hu MY. 1998. Forecasting with Artificial Neural Networks: The State of The Art. International Journal of Forecasting, 14:35–62. https://doi.org/10.1016/S0169-2070(97)00044-7
  • Zhang W, Chen Q, Yan J, Zhang S, Xu J. 2021. A novel asynchronous deep reinforcement learning model with adaptive early forecasting method and reward incentive mechanism for short-term load forecasting. Energy, 236:e:121492. https://doi.org/10.1016/j.energy.2021.121492

FORECASTING TURKEY ELECTRICITY CONSUMPTION WITH DEEP LEARNING BI-LSTM MODEL*

Year 2022, Volume: 1 Issue: 1, 24 - 33, 31.12.2022

Abstract

The energy consumption of Turkey, which is among the developing countries, is constantly increasing. Despite this increasing energy need, it is an insufficient country in terms of energy production. Turkey, which is a foreign-dependent country in energy use, has problems with sustainable energy supply. Especially recently, Russia's restrictions on energy exports to European countries have caused an energy crisis all over the world. For this reason, energy supply security has a vital role for Turkey as well as for the rest of the world. In this context, the estimation of energy consumption for future periods is a strategic issue that should be emphasized. In the study, monthly energy consumption amounts of Turkey between January 2005 and November 2018 were taken and a five-year estimate of the ever-increasing electricity consumption in the range of 2019-2023 was made using bi-directional LSTM models (ADAM, RmsProp, SGDM). The highest performance in the models was obtained with RMSprop optimization. The monthly electrical energy consumption data between 2019-2020 and the estimated data of monthly electricity consumption for the same period obtained by RMSprop optimization were compared. According to the optimization result, Turkey's electricity consumption will continue to increase. Turkey should put into effect the necessary plans quickly in the face of this increasing need. Incorporating the education of households into plans for energy conservation may be a viable solution.

References

  • Ahmad AS, Hassan MY, Abdullah MP, Rahman HA, Hussin F, Abdullah H, Saidur R. 2014. A review on applications of ANN and SVM for building electrical energy consumption forecasting. Renewable and Sustainable Energy Reviews, 33:102-109. https://doi.org/10.1016/j.rser.2014.01.069
  • Anand A, Suganthi L. 2017. Forecasting of electricity demand by hybrid ANN-PSO models. International Journal of Energy Optimization and Engineering, 6(4):66-80. doi:10.4018/IJEOE.2017100105
  • EPDK. 2021. Electricity Market Sector Report. Ankara.
  • EPDK. 2020. Electricity Market Sector Report. Ankara.
  • EPDK, 2019. Electricity Market Sector Report, Ankara.
  • Fara L, Diaconu A, Craciunescu D, Fara S. 2021. Forecasting of energy production for photovoltaic systems based on ARIMA and ANN advanced models. International Journal of Photoenergy, 2021: e:6777488. https://doi.org/10.1155/2021/6777488
  • Cui Z, Ke R, Wang Y. 2017. Deep Stacked Bidirectional and Unidirectional LSTM Recurrent Neural Network for Network-wide Traffic Speed Prediction. Available from: https://arxiv.org/ftp/arxiv/papers/1801/1801.02143.pdf
  • Dong K, Dong X, Jiang Q. 2020. How renewable energy consumption lower global CO2 emissions? Evidence from countries with different income levels. The World Economy, 43:1665-1698. doi: 10.1111/twec.12898
  • Enerdata. 2022. https://yearbook.enerdata.net/total-energy/world-consumption-statistics.html
  • Erilli NA, Eğrioğlu E,Yolcu U, Aladağ HÇ, Uslu VR. 2010. Türkiye’de Enflasyonun İleri ve Geri Beslemeli Yapay Sinir Ağlarının Melez Yaklaşımı ile Öngörüsü. Doğuş Üniversitesi Dergisi, 11(1):42-55. Available from: https://dergipark.org.tr/tr/pub/doujournal/issue/66662/1042993
  • Fauset L. 1994. Fundamentals of Neural Network. Prentice Hall International, London
  • Gandelli A, Grimaccia F, Leva S, Mussetta M, Ogliari E. 2014. Hybrid model analysis and validation for PV energy production forecasting. 2014 International Joint Conference on Neural Networks (IJCNN), July 6-11, China. Available from: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6889786
  • Gref K, Srivastava RK, Koutnik J, Steunebrink BR, Schmindhuber J. 2017. LSTM-A Search Space Odysse, IEEE Transactions on Neural Networks and Learning Systems, 28(10):2222-2232. doi: 10.1109/TNNLS.2016.2582924
  • Haykin S. 1998. Neural Networks: A Comprehensive Foundation. Seconf Ed., Prentice hall, New Jersey.
  • Koukaras P, Bezas N, Gkaidatzis P, Ionnidis D, Tzovaras D, Tjortjis C. 2021. Introducing a novel approach in one-step ahead energy load forecasting. Sustainable Computing: Informatics and Systems, 32:100616. https://doi.org/10.1016/j.suscom.2021.100616
  • Li K, Su H, Chu J. 2011. Forecasting building energy consumption using neural networks and hybrid neuro-fuzzy system: A comparative study. Energy and Buildings, 43:2893-2899. doi:10.1016/j.enbuild.2011.07.010
  • Pazikadin RA, Rifai D, Ali K, Malik MZ, Abdalla A N, Faraj MA. 2020. Solar irradiance measurement instrumentation and power solar generation forecasting based on Artificial Neural Networks (ANN): A review of five years research trend. Science of the Total Environment, 715:e:136848. https://doi.org/10.1016/j.scitotenv.2020.136848
  • TEİAŞ. 2022. Available from: https://www.teias.gov.tr/turkiye-elektrik-uretim-iletim-istatistikleri
  • Zhang G, Patuwo BE, Hu MY. 1998. Forecasting with Artificial Neural Networks: The State of The Art. International Journal of Forecasting, 14:35–62. https://doi.org/10.1016/S0169-2070(97)00044-7
  • Zhang W, Chen Q, Yan J, Zhang S, Xu J. 2021. A novel asynchronous deep reinforcement learning model with adaptive early forecasting method and reward incentive mechanism for short-term load forecasting. Energy, 236:e:121492. https://doi.org/10.1016/j.energy.2021.121492
There are 20 citations in total.

Details

Primary Language English
Subjects Engineering
Journal Section Research Articles
Authors

Hatice Genç Kavas 0000-0002-6813-529X

Publication Date December 31, 2022
Published in Issue Year 2022 Volume: 1 Issue: 1

Cite

APA Genç Kavas, H. (2022). FORECASTING TURKEY ELECTRICITY CONSUMPTION WITH DEEP LEARNING BI-LSTM MODEL*. Sivas Cumhuriyet Üniversitesi Bilim Ve Teknoloji Dergisi, 1(1), 24-33.
AMA Genç Kavas H. FORECASTING TURKEY ELECTRICITY CONSUMPTION WITH DEEP LEARNING BI-LSTM MODEL*. CUJAST. December 2022;1(1):24-33.
Chicago Genç Kavas, Hatice. “FORECASTING TURKEY ELECTRICITY CONSUMPTION WITH DEEP LEARNING BI-LSTM MODEL*”. Sivas Cumhuriyet Üniversitesi Bilim Ve Teknoloji Dergisi 1, no. 1 (December 2022): 24-33.
EndNote Genç Kavas H (December 1, 2022) FORECASTING TURKEY ELECTRICITY CONSUMPTION WITH DEEP LEARNING BI-LSTM MODEL*. Sivas Cumhuriyet Üniversitesi Bilim ve Teknoloji Dergisi 1 1 24–33.
IEEE H. Genç Kavas, “FORECASTING TURKEY ELECTRICITY CONSUMPTION WITH DEEP LEARNING BI-LSTM MODEL*”, CUJAST, vol. 1, no. 1, pp. 24–33, 2022.
ISNAD Genç Kavas, Hatice. “FORECASTING TURKEY ELECTRICITY CONSUMPTION WITH DEEP LEARNING BI-LSTM MODEL*”. Sivas Cumhuriyet Üniversitesi Bilim ve Teknoloji Dergisi 1/1 (December 2022), 24-33.
JAMA Genç Kavas H. FORECASTING TURKEY ELECTRICITY CONSUMPTION WITH DEEP LEARNING BI-LSTM MODEL*. CUJAST. 2022;1:24–33.
MLA Genç Kavas, Hatice. “FORECASTING TURKEY ELECTRICITY CONSUMPTION WITH DEEP LEARNING BI-LSTM MODEL*”. Sivas Cumhuriyet Üniversitesi Bilim Ve Teknoloji Dergisi, vol. 1, no. 1, 2022, pp. 24-33.
Vancouver Genç Kavas H. FORECASTING TURKEY ELECTRICITY CONSUMPTION WITH DEEP LEARNING BI-LSTM MODEL*. CUJAST. 2022;1(1):24-33.