SavedAdd your journal

Journals / Verimlilik Dergisi / 2023 / Cilt: 57 - Sayı: 1

Forecasting of Renewable Energy Generation for Turkey by Artificial Neural Networks and ARIMA Model: 2023 Generation Targets by Renewable Energy Resources

Yapay Sinir Ağları ve ARIMA Modeli ile Türkiye Için Yenilenebilir Enerji Üretiminin Tahmini: 2023 Yenilenebilir Enerji Kaynaklarına Göre Üretim Hedefleri

Özlem KARADAĞ ALBAYRAK
Pages
121–138
Publication Date
2023-01-31
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

These results show that the ANN model gives a more appropriate estimation result.Originality: In this research, a new model was proposed for the amount of energy to be obtained from RES in Türkiye.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Purpose: Türkiye attaches particular importance to the energy production with renewable energy sources in order to overcome the negative economic, environmental and social effects which are caused by fossil resources in energy production.
📊
Key Findings & Data:The aim of this study is to propose a model for forecasting the amount of energy to be produced for Türkiye using renewable energy resources.Methdology: In this study, a forecasting model was created by using the generatio amount of energy generation from renewable sources data between 1965 and 2019 and by using Artificial Neural Networks (ANN) and Autoregressive Integrated Moving Average (ARIMA) methods.Findings: While it was estimated that 127.516 TWh of energy will be produced in 2023 with the ANN method, this amount was estimated as 45,457 TeraWatt Hours (TWh) with the ARIMA (1,1,6) model.
💡
Conclusion & Impact:These results show that the ANN model gives a more appropriate estimation result.Originality: In this research, a new model was proposed for the amount of energy to be obtained from RES in Türkiye.
🔬Design:Modeling & Simulation
📍Scope:2019 Verileri
⚙️Variables:Enerji · Yenilenebilir Enerji Kaynakları · Yenilenebilir Enerji Üretimi Tahmini

Abstract

Purpose: Türkiye attaches particular importance to the energy production with renewable energy sources in order to overcome the negative economic, environmental and social effects which are caused by fossil resources in energy production. The aim of this study is to propose a model for forecasting the amount of energy to be produced for Türkiye using renewable energy resources.Methdology: In this study, a forecasting model was created by using the generatio amount of energy generation from renewable sources data between 1965 and 2019 and by using Artificial Neural Networks (ANN) and Autoregressive Integrated Moving Average (ARIMA) methods.Findings: While it was estimated that 127.516 TWh of energy will be produced in 2023 with the ANN method, this amount was estimated as 45,457 TeraWatt Hours (TWh) with the ARIMA (1,1,6) model. Mean Absolute Percent Error (MAPE) was calculated in order to determine the margin of error of the forecasting models. These values were determined as 13.1% for the ANN model and 21.9% for the ARIMA model. These results show that the ANN model gives a more appropriate estimation result.Originality: In this research, a new model was proposed for the amount of energy to be obtained from RES in Türkiye. It is thought that the results obtained in this study will be useful in energy planning and management.

Özet

Amaç: Türkiye, enerji üretiminde fosil kaynakların neden olduğu olumsuz ekonomik, çevresel ve sosyal etkileri ortadan kaldırmak için yenilenebilir enerji kaynakları ile enerji üretimine ve tahminine çok önem vermektedir. Bu çalışmanın amacı, Yenilenebilir enerji kaynakları kullanılarak Türkiye’de üretecek enerji miktarını tahmin etmek için bir model önermektir.Yöntem: Bu araştırma 1965-2019 yılları arasında yenilenebilir kaynaklı enerji üretim verileri kullanılarak Yapay Sinir Ağları (YSA) ve Otoregresif Bütünleşik Hareketli Ortalama (ARIMA) yöntemlerinden yararlanılmıştır.Bulgular: ANN yönteminde, 2023 yılında 127.516 TWh enerji üretileceği tahminedilirken, ARIMA (1.1.6) modeli ile bu miktarın 45.457 TeraWatt Saat (TWh) olacağı tahmin edilmiştir. Tahmin modellerinin hata payını belirlemek için Ortalama Mutlak Yüzde Hatası (MAPE) hesaplanmıştır. Bu değer YSA modeli ile %13,1, ARIMA modeli ile %21,9 olarak belirlenmiştir. YSA modelinin daha doğru bir sonuç verdiğini göstermiştir.Özgünlük: Türkiye’de YEK’dan elde edilecek enerji miktarı tahmini için model önerisi yapılmıştır. Bu çalışmada ulaşılan sonuçların enerji planlaması ve yönetiminde faydalı olacağı düşünülmektedir.

🔖

Cite This Article

ALBAYRAK, Ö. K. (2023). Forecasting of Renewable Energy Generation for Turkey by Artificial Neural Networks and ARIMA Model: 2023 Generation Targets by Renewable Energy Resources. Verimlilik Dergisi, Cilt: 57 - Sayı: 1, 121–138. https://ajindex.com/articles/880281

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (11)

  1. Abdirassilov, Z., Sładkowski, A. (2018). “Application of Artificial Neural Networks for Shortterm Prediction of Container Train Flows in Direction of China-Europe via Kazakhstan”, Transport Problems, 13(4): 103-113.
  2. Ahmad Tzhanga, H., Yana, B. (2020). “A Review on Renewable Energy and Electricity Requirement Forecasting Models for Smart Grid and Buildings”, Sustainable Cities and Society, 55, 102052.
  3. Alkan, O., Albayrak, O.K. (2020), “Ranking of Renewable Energy Sources for Regions in Turkey by Fuzzy Entropy Based Fuzzy COPRAS and Fuzzy MULTIMOORA”, Renewable Energy, 162, 712-726. https://doi.org/10.1016/j.renene.2020.08.062.
  4. Araujo da Silva Junior, J.C., Michaelsen, A.L., Scalvi, M., Pacheco, M.T.G. (2018). “ Forecast of Electric Energy Generation Potential from Swine Manure in Santa Catarina, Brazil, Environment”, Development and Sustainability, 22 (3): 2305-2319.
  5. Asensio, J.J., Darado, F., Duran, J. (2020). “Energy Demand Forecasting Using Deep Learning: Applications for the French Grid”, Energies, 13: 2242-2257.
  6. Box, G.E.P., Jenkins, G.M. (1976). sTime series analysis. Forecasting and Controls. Revised edition. San Francisco: Holden Day. Bhardwaj, S., Chandrasekhar. E., Padiyar, P., Gadre, V.M. (2020). “A Comparative Study of Wavelet-Based ANN and Classical Techniques for Geophysical Time-Series Forecasting”, Computers & Geosciences, 138, 104461. BP (2020). “Statistical Review of World Energy 2020”, https://www.bp.com/content/dam/bp/business-sites/en/global/corporate/pdfs/energy-economics/statistical-review/bp-stats-review-2020-full-report.pdf, (Access Date: 01.10.2020)
  7. Broadny, J., Tutak. M., Saki, S.A. (2020). “Forecasting the Structure of Energy Production from Renewable Energy Sources and Biofuels in Poland”, Energies, 13, 2539-2570.
  8. Cengiz, C., Manga, M. (2021).” The Causal Linkages between Renewable Energy Consumption, Economic Growth, Oil Prices and CO2 Emissions in Selected OECD Countries”, Verimlilik Dergisi, 3, 165-183.
  9. Cinar, D., Kayakutlu, G. (2007). “Forecasting Production of Renewable Energy Using Cognitive Mapping and Artificial Neural Networks”, 19th International Conference on Production Research, Proceedings, Valparaiso, Chile Ediger, V.S., Akar, S. (2007). “ARIMA Forecasting of Primary Energy Demand by Fuel in Turkey”, Energy Policy, 35, 1701-1708.
  10. EPDK (2020). “Electricity Market Annual Sector Report”, https://www.epdk.gov.tr/Detay/Icerik/3-0-24/elektrikyillik-se%22ktor-raporu, (Access Date: 20.12.2022).
  11. Erdoğdu, E. (2007). “Electr