Türkiye’nin Enerji Talebi Tahmin Probleminin Çözümünde Regresyon Yöntemlerine Dayalı Yaklaşımlar

Öz

Günümüzde enerji talebi ve bu talebin gelecekteki tahmini, sürdürülebilir enerji politikaları oluşturmak ve enerji kaynaklarının etkin kullanımını sağlamak için hayati bir öneme sahiptir. Bu çalışmada Türkiye’nin 1979-2020 yılları arasına ait gözlemlenen enerji talebi, nüfus, gayri safi yurt içi hasıla, ihracat ve ithalat verileri kullanılarak enerji talep tahmin modelleri oluşturulmuştur. Enerji talep tahmini modellerini oluşturmak için çoklu regresyon ve polinom regresyon yöntemleri kullanılmıştır. Çalışmanın temel hedefi, literatürde bulunan yöntemlere kıyasla doğruluk oranı daha fazla olan bir talep tahmin modeli sunmaktır. Çalışmanın başlıca bulguları, çoklu regresyonun enerji talep tahmininde literatürdeki çalışmaların sonucu yakaladığını ve etkili bir araç olduğunu göstermektedir. Ayrıca, polinom regresyon tabanlı yaklaşımla literatürdeki sonuçlar yaklaşık %4 oranında iyileştirilmiş ve güvenli bir tahmin modeli literatüre sunulmuştur. Ayrıca, elde edilen tahmin modeli kullanılarak Türkiye’nin 2021-2050 arasındaki enerji talebi tahmin edilmiştir. Elde edilen sonuçlar, belirtilen dönemde enerji talebinin önemli ölçüde artacağını ortaya koymaktadır. 2021-2050 yılları arası enerji talebi artış oranının, geçmiş yıllara benzerlik göstermesi, regresyon tabanlı yöntemin güvenilirliğini desteklemektedir. Bu çalışma, enerji planlaması ve politika oluşturmadaki karar alıcılar için önemli bir referans kaynağı olabilir.

Anahtar Kelimeler

• Enerji talep tahmini
• regresyon yöntemleri
• polinom regresyon.

Approaches Based on Regression Methods in Solving Turkey’s Energy Demand Forecasting Problem

Öz

Today’s energy demand and the future prediction of this demand are of vital importance to create sustainable energy policies and ensure the effective use of energy resources. In this study, energy demand forecast models were created using Turkey’s observed energy demand, population, gross domestic product, export and import data between 1979 and 2020. Multiple regression and polynomial regression methods were used to create energy demand forecast models. The main goal of the study is to present a demand forecasting model with a higher accuracy rate compared to the methods found in the literature. The main findings of the study show that multiple regression captures the results of studies in the literature and is an effective tool in energy demand forecasting. In addition, with the polynomial regression-based approach, the results in the literature were improved by approximately 4% and a reliable prediction model was presented to the literature. Additionally, Turkey’s energy demand between 2021-2050 was estimated using the resulting forecast model. The results obtained reveal that energy demand will increase significantly in the specified period. The fact that the energy demand increase rate between 2021 and 2050 is similar to previous years supports the reliability of the regression-based method. This study can be an important reference source for decision makers in energy planning and policy making.

Anahtar Kelimeler

• Energy demand forecasting
• regression methods
• polynomial regression.

Kaynakça

1. Kahraman G. Türkiye’de kentleşmenin enerji tüketimi ve karbon salınımı üzerine etkisi. Journal of the Institute of Science and Technology 2019; 9(3): 1559-1566.
2. Nia AR, Awasthi A, Bhuiyan N. Industry 4.0 and demand forecasting of the energy supply chain: A literature review’. Comput Ind Eng 2021; 154: 107128.
3. Ozili PK, Ozen E. Global Energy Crisis: Impact on the Global Economy. In: Sood K, Grima S, Young P, Ozen E, Balusamy B, editors. The Impact of Climate Change and Sustainability Standards on the Insurance Market. 1st ed. Wiley; 2023. pp. 439-454.
4. Zakeri B, et al. Pandemic, war, and global energy transitions. Energies 2022; 15(17): 6114.
5. Peng B, Chang BH, Yang L, Zhu C. Exchange rate and energy demand in G7 countries: Fresh insights from Quantile ARDL model. Energy Strategy Rev 2022; 44: 100986.
6. Taghvaee VM, Nodehi M, Assari Arani A, Rishehri M, Nodehi SE, Shirazi JK. Fossil fuel price policy and sustainability: energy, environment, health and economy. Int. J Energy Sect. Manag 2023; 17(2): 371-409.
7. Astrov V, Hanzl-Weiss D, Leitner SM, Pindyuk O, Pöschl J, Stehrer R. Energy efficiency and EU industrial competitiveness: Energy costs and their impact on manufacturing activity. The Vienna Institute for International Economic Studies Research Report. Vienna, Austria; 2015.
8. Zhang J, Tan Z, Wei Y. An adaptive hybrid model for short term electricity price forecasting. Appl Energy 2020; 258: 114087.

9. Beskirli M, Hakli H, Kodaz H. The energy demand estimation for Turkey using differential evolution algorithm. Sādhanā. 2017; 42(10): 1705-1715.
10. Rafique SF, Jianhua Z. Energy management system, generation and demand predictors: a review. IET Gener Transm Distrib 2018; 12(3): 519-530.
11. Toksarı MD. Ant colony optimization approach to estimate energy demand of Turkey. Energy Policy 2007; 35(8): 3984-3990.
12. Rao C, Zhang Y, Wen J, Xiao X, Goh M. Energy demand forecasting in China: A support vector regression-compositional data second exponential smoothing model. Energy 2023; 263: 125955.
13. Avtar R, Tripathi S, Aggarwal AK, Kumar P. Population–urbanization–energy nexus: a review. Resources. 2019; 8(3): 136.
14. Supersberger N, Führer L. Integration of renewable energies and nuclear power into North African Energy Systems: An analysis of energy import and export effects. Energy Policy 2011; 39(8): 4458-4465.
15. Stern DI. Energy-GDP relationship. The New Palgrave Dictionary of Economics; Palgrave Macmillan: London, UK; 2018.
16. Liu W-C. The relationship between primary energy consumption and real gross domestic product: Evidence from major Asian countries. Sustainability 2020; 12(6): 2568.
17. Sheffield J. World population and energy demand growth: the potential role of fusion energy in an efficient world. Philos. Trans. R. Soc. London, Ser A 1999; 357(1752): 377-395.
18. Dedeoğlu D, Kaya H. Energy use, exports, imports and GDP: New evidence from the OECD countries. Energy Policy 2013; 57: 469-476.
19. Carfora A, Pansini RV, Scandurra G. Energy dependence, renewable energy generation and import demand: Are EU countries resilient? Renewable Energy. 2022; 195: 1262-1274.
20. Shahzad U, Doğan B, Sinha A, Fareed Z. Does Export product diversification help to reduce energy demand: Exploring the contextual evidences from the newly industrialized countries. Energy 2021; 214: 118881.
21. Ahmad T, Zhang D. A critical review of comparative global historical energy consumption and future demand: The story told so far. Energy Rep 2020; 6: 1973-1991.
22. Beşkirli A, Temurtaş H, Özdemir D. Determination with Linear Form of Turkey’s Energy Demand Forecasting by the Tree Seed Algorithm and the Modified Tree Seed Algorithm. Advances in Electrical & Computer Engineering, 2020; 20(2).
23. Özdemir D, Dörterler S, Aydın D. A new modified artificial bee colony algorithm for energy demand forecasting problem. Neural Comput Appl 2022; 34(20): 17455-17471.
24. Özdemir D, Dörterler S. An adaptive search equation-based artificial bee colony algorithm for transportation energy demand forecasting. Turk J Electr Eng Comput Sci 2022; 30(4): 1251-1268.
25. Bilici Z, Özdemir D, Temurtaş H. Comparative analysis of metaheuristic algorithms for natural gas demand forecasting based on meteorological indicators. J Eng Res 2023; 11(3): 259-265.
26. Chatterjee S, Hadi AS. Regression analysis by example. John Wiley & Sons, USA; 2013.
27. Draper NR, Smith H. Applied regression analysis. John Wiley & Sons, USA; 1998.
28. Narin S, Doğan O, Bande N, Yunus G. Keçiören/Ankara Özelinde Konut Rayiç Değerlerinin Tahmininde Çoklu Regresyon Analizi ve Yapay Sinir Ağları Yöntemlerinin Karşılaştırılması. International Journal of Engineering Research and Development 2023; 15(2): 828-839.
29. Sajid T, Jamshed W, Ibrahim RW, Eid MR, Abd-Elmonem A, Arshad M. Quadratic regression analysis for nonlinear heat source/sink and mathematical Fourier heat law influences on Reiner-Philippoff hybrid nanofluid flow applying Galerkin finite element method. Journal of Magnetism and Magnetic Materials 2023; 568: 170383.
30. Lin P, Hong Y, He Y, Pei M. Advancing and lagging effects of weather conditions on intercity traffic volume: A geographically weighted regression analysis in the Guangdong-Hong Kong-Macao Greater Bay Area. International Journal of Transportation Science and Technology 2024; 13: 58-76.
31. Kim D-H, et al. Regression analysis of nasal shape from juvenile to adult ages for forensic facial reconstruction. Legal Med 2024; 66: 102363.
32. Vardopoulos I. Adaptive Reuse for Sustainable Development and Land Use: A Multivariate Linear Regression Analysis Estimating Key Determinants of Public Perceptions. Heritage 2023; 6(2): 809-828.
33. Arkes J. Regression analysis: a practical introduction. Taylor & Francis, England; 2019.
34. Sykes AO. An introduction to regression analysis. USA; 1993.
35. Aslan M. Archimedes optimization algorithm based approaches for solving energy demand estimation problem: a case study of Turkey. Neural Comput Appl 2023; 35(26): 19627-19649.
36. Toğaçar M. Arşimet Optimizasyon Algoritması ile Trafo Tabanlı Evrişimsel Sinir Ağı Modelini Kullanarak Yazılım Tanımlı Ağ Teknolojisi Verilerinde Saldırı Tespiti. Fırat Üniversitesi Mühendislik Bilimleri Dergisi 2022; 34(1): 341-349.
37. Hashim FA, Hussain K, Houssein EH, Mabrouk MS, Al-Atabany W. Archimedes optimization algorithm: a new metaheuristic algorithm for solving optimization problems. Appl Intell 2021; 51(3): 1531-1551.
38. Pham D, Karaboga D. Intelligent optimisation techniques: genetic algorithms, tabu search, simulated annealing and neural networks. Springer Science & Business Media, Germany; 2012.
39. Karaboğa D. Yapay zeka optimizasyon algoritmaları. Nobel Akademi Yayıncılık, Türkiye; 2014.