Hastaneler için Yenilenebilir Enerji Yol Haritası: Türkiye’de Makine Öğrenimi Destekli Hibrit Güneş-Rüzgâr Sistemi Tahmini ve Boyutlandırması

Abstract

Binalarda ısıtma, soğutma ve aydınlatma amaçlı enerji kullanımı artmaya devam ederek fosil yakıt tüketimini ve karbon emisyonlarını artırmaktadır. Hastaneler özellikle yüksek bir taleple karşı karşıyadır. Bu çalışma, Türkiye’nin Diyarbakır kentinde hastane ölçeğinde bir hibrit yenilenebilir enerji sistemi (güneş PV ve rüzgâr) tasarlayıp değerlendirerek, enerji tüketimini azaltmak için gerçek verilere dayalı performans modelleri geliştirmektedir. PV ve rüzgâr enerjisi üretimini tahmin etmek için hava durumu ve bina özelliklerini içeren Pace regresyonu kullanılmıştır. Üç aylık veriler (Ağustos–Ekim 2022) — ortalama sıcaklık 26 °C, en yüksek güneş ışınımı 1012 W/m², ortalama rüzgar hızı 1,18 m/s — ve toplam 2895 MWh elektrik tüketimi analiz edilmiştir. Modeller, RMSE ve MAPE kullanılarak doğrulanmıştır. Sonuçlar, güçlü bir uyum göstermektedir (PV: R² = 0,9262, RMSE = 0,037; rüzgâr: R² = 0,986, RMSE = 0,021). Tahmini hatalar güneş enerjisi için %14, rüzgar enerjisi için %11 olarak hesaplanmıştır. Senaryo analizi, hibrit sistemin hastanenin yıllık talebinin yaklaşık %7,14'ünü PV yoluyla ve %0,02'sini rüzgar yoluyla karşılayabileceğini göstermektedir. Rüzgarın zayıf olduğu koşullarda bile, makine öğrenimi destekli tasarım tahmin ve boyutlandırmayı iyileştirerek, çeşitli iklimlerdeki kamu binaları için yenilenebilir enerji kaynaklarını optimize etmek, şebekeye bağımlılığı azaltmak ve hizmet güvenilirliğini korurken emisyonları düşürmek için pratik bir yol haritası sunmaktadır.

Keywords

• bina enerji tüketimi
• güneş enerjisi
• yenilenebilir enerji sistemi tasarımı
• yapay zeka

Renewable Energy Roadmap for Hospitals: ML-Enabled Forecasting and Sizing of a Hybrid PV–Wind System in Türkiye

Abstract

Energy use for heating, cooling, and lighting in buildings continues to increase, driving fossil fuel consumption and carbon emissions. Hospitals face especially high demand. This study designs and evaluates a hospital-scale hybrid renewable system (solar PV plus wind) for Diyarbakır, Türkiye, and develops real data-driven performance models to reduce energy consumption. Pace regression, incorporating weather and building characteristics, is used to predict PV and wind generation. Three months of data (Aug–Oct 2022)—average temperature 26 °C, peak solar irradiance 1012 W/m², average wind speed 1.18 m/s—and total electricity use of 2895 MWh were analyzed. Models were validated using RMSE and MAPE. Results show strong fit (PV: R² = 0.9262, RMSE = 0.037; wind: R² = 0.986, RMSE = 0.021). Estimated errors were 14% for solar and 11% for wind. Scenario analysis indicates the hybrid system could supply approximately 7.14% of the hospital’s annual demand via PV and 0.02% via wind. Even in modest-wind regimes, machine-learning-guided design enhances forecasting and sizing, providing a practical roadmap for public buildings in diverse climates to optimize renewables, reduce grid dependence, and lower emissions while maintaining service reliability.

Keywords

• building energy consumption
• solar energy
• renewable energy system design
• artificial intelligence

Ethical Statement

Bu çalışma için etik kurul izni gerekmemektedir. Yazarlar, çalışmanın hazırlanması sürecinde araştırma ve yayın etiği ilkelerine uyulduğunu beyan eder.

References

1. [1] D. Ürge-Vorsatz, L.F. Cabeza, S. Serrano, C. Barreneche, K. Petrichenko, Heating and cooling energy trends and drivers in buildings, Renewable and Sustainable Energy Reviews. 41 (2015) 85–98.
2. [2] M.H. Jahangir, S. Eslamnezhad, S.A. Mousavi, M. Askari, Multi-year sensitivity evaluation to supply prime and deferrable loads for hospital application using hybrid renewable energy systems, Journal of Building Engineering. 40 (2021) 102733.
3. [3] A. González González, J. García-Sanz-Calcedo, D. Rodríguez Salgado, Evaluation of energy consumption in German hospitals: Benchmarking in the public sector, Energies. 11 (2018) 2279.
4. [4] M. Economidou, V. Todeschi, P. Bertoldi, D. D’Agostino, P. Zangheri, L. Castellazzi, Review of 50 years of EU energy efficiency policies for buildings, Energy and Buildings. 225 (2020) 110322.
5. [5] K. Chau, K. Mahani, M.A. Jafari, S. Haghani, Solar-powered microgrid capacity planning for a general hospital, in: IEEE, 2018: pp. 1–6.
6. [6] D.M. Alotaibi, M. Akrami, M. Dibaj, A.A. Javadi, Smart energy solution for an optimised sustainable hospital in the green city of NEOM, Sustainable Energy Technologies and Assessments. 35 (2019) 32–40.
7. [7] S. Okamoto, Saving Energy in Hospitals with CCHP, in: 2011: p. 126.
8. [8] Y. He, Q. Gong, Z. Zhou, H. Chen, Development of a hybrid VRF system energy consumption prediction model based on data partitioning and swarm intelligence algorithm, Journal of Building Engineering. 74 (2023) 106868.

9. [9] S.M. Vaziri, B. Rezaee, M.A. Monirian, Utilizing renewable energy sources efficiently in hospitals using demand dispatch, Renewable Energy. 151 (2020) 551–562.
10. [10] S. Mandal, B.K. Das, N. Hoque, Optimum sizing of a stand-alone hybrid energy system for rural electrification in Bangladesh, Journal of Cleaner Production. 200 (2018) 12–27.
11. [11] C. Ammari, D. Belatrache, B. Touhami, S. Makhloufi, Sizing, optimization, control and energy management of hybrid renewable energy system—A review, Energy and Built Environment. 3 (2022) 399–411.
12. [12] M.G. Gomes, A.M. Rodrigues, F. Natividade, Thermal and energy performance of medical offices of a heritage hospital building, Journal of Building Engineering. 40 (2021) 102349.
13. [13] L. Cao, Y. Li, J. Zhang, Y. Jiang, C. Han, J. Wei, Electrical load prediction of healthcare buildings through single and ensemble learning, Energy Reports. 6 (2020) 2751–2767
14. [14] M. Fatehijananloo, H. Stopps, J.J. McArthur, Exploring artificial intelligence methods for energy prediction in healthcare facilities: An in-depth extended systematic review, Energy and Buildings. 320 (2024) 114598.
15. [15] M. As, T. Bilir, Machine learning algorithms for energy efficiency: Mitigating carbon dioxide emissions and optimizing costs in a hospital infrastructure, Energy and Buildings. 318 (2024) 114494.
16. [16] C. Correa-Jullian, J.M. Cardemil, E.L. Droguett, M. Behzad, Assessment of deep learning techniques for prognosis of solar thermal systems, Renewable Energy. 145 (2020) 2178–2191.
17. [17] J. Yan, C. Tian, J. Huang, Y. Wang, Load forecasting using twin gaussian process model, in: IEEE, 2012: pp. 36–41.
18. [18] L. Zhang, J. Wen, Y. Li, J. Chen, Y. Ye, Y. Fu, W. Livingood, A review of machine learning in building load prediction, Applied Energy. 285 (2021) 116452.
19. [19] I. Soyler, E. Izgi, Electricity demand forecasting of hospital buildings in Istanbul, Sustainability. 14 (2022) 8187.
20. [20] H.-F. Shi, The hourly load forecasting based on linear Gaussian state space model, in: IEEE, 2012: pp. 741–747.
21. [21] A. Khosravi, S. Nahavandi, Load forecasting using interval type-2 fuzzy logic systems: Optimal type reduction, IEEE Transactions on Industrial Informatics. 10 (2013) 1055–1063.
22. [22] J.G. Jetcheva, M. Majidpour, W.-P. Chen, Neural network model ensembles for building-level electricity load forecasts, Energy and Buildings. 84 (2014) 214–223.
23. [23] V. RICHALET, F. NEIRAC, F. TELLEZ, J. MARCO, J.J. BLOEM, HELP (House Labelling Procedure): Methodology and Present Results., (n.d.).
24. [24] A. Zhao, M. Chen, W. Quan, S. Zhang, A hybrid forecasting model for general hospital electricity consumption based on mixed signal decomposition, Energy and Buildings. 325 (2024) 115006.
25. [25] R. Sendra-Arranz, A. Gutiérrez, A long short-term memory artificial neural network to predict daily HVAC consumption in buildings, Energy and Buildings. 216 (2020) 109952.
26. [26] A. Bagnasco, F. Fresi, M. Saviozzi, F. Silvestro, A. Vinci, Electrical consumption forecasting in hospital facilities: An application case, Energy and Buildings. 103 (2015) 261–270.
27. [27] A. Bagnasco, F. Fresi, M. Saviozzi, F. Silvestro, A. Vinci, Electrical consumption forecasting in hospital facilities, in: Machine Learning and Data Science in the Power Generation Industry, Elsevier, 2021: pp. 195–211.
28. [28] M. Zini, C. Carcasci, Machine learning-based monitoring method for the electricity consumption of a healthcare facility in Italy, Energy. 262 (2023) 125576.
29. [29] I. Santiago, D. Trillo-Montero, I. Moreno-Garcia, V. Pallarés-López, J. Luna-Rodríguez, Modeling of photovoltaic cell temperature losses: A review and a practice case in South Spain, Renewable and Sustainable Energy Reviews. 90 (2018) 70–89.
30. [30] https://sd-windenergy.com/, n.d.
31. [31] https://wind-data.ch/tools/profile.php?lng=en, n.d.
32. [32] B. Bilal, K.H. Adjallah, K. Yetilmezsoy, M. Bahramian, E. Kıyan, Determination of wind potential characteristics and techno-economic feasibility analysis of wind turbines for Northwest Africa, Energy. 218 (2021) 119558.
33. [33] A. Tazay, Techno-economic feasibility analysis of a hybrid renewable energy supply options for university buildings in Saudi Arabia, Open Engineering. 11 (2020) 39–55.
34. [34] Y. Wang, I.H. Witten, Pace Regression, Computer Science Working Papers 99/12, University of Waikato, Hamilton, New Zealand, 1999.
35. [35] Y. Wang, I.H. Witten, Modeling for optimal probability prediction, (2002).
36. [36] S. Gyamfi, B. Aboagye, M. Obeng, F. Peprah, Criticality and severity of adverse effects of the sun on performance of solar PV systems, Sol. Energy Adv. 4 (2024) 100058.
37. [37] I.K. Okakwu, A.S. Alayande, D.O. Akinyele, O.E. Olabode, J.O. Akinyemi, Effects of total system head and solar radiation on the techno-economics of PV groundwater pumping irrigation system for sustainable agricultural production, Sci. African 16 (2022) e01118.
38. [38] K. Mertens, photovoltaic technology Technology and Practice, 2020.
39. [39] H. Goverde, D. Goossens, J. Govaerts, F. Catthoor, K. Baert, J. Poortmans, J. Driesen, Spatial and temporal analysis of wind effects on PV modules: Consequences for electrical power evaluation, Sol. Energy 147 (2017) 292–299.
40. [40] M.S. Ahmed, R. Karal, B.K. Das, A. Das, Experimental investigation of cooling, wind velocity, and dust deposition effects on solar PV performance in a tropical climate in Bangladesh, Case Stud. Therm. Eng. 50 (2023).
41. [41] E. Simsek, M.J. Williams, L. Pilon, Effect of dew and rain on photovoltaic solar cell performances, Sol. Energy Mater. Sol. Cells 222 (2021) 110908.
42. [42] R. Ahmed, V. Sreeram, Y. Mishra, M.D. Arif, A review and evaluation of the state-of-the-art in PV solar power forecasting: Techniques and optimization, Renew. Sustain. Energy Rev. 124 (2020).
43. [43] A. Ataman, N. Arslanoğlu, Investigation of the cooling effect of wind on rooftop PV power plants, Case Stud. Therm. Eng. 63 (2024) 1–15.
44. [44] M. Mehdi, N. Ammari, A. Alami Merrouni, A. Benazzouz, M. Dahmani, Experimental investigation on the effect of wind as a natural cooling agent for photovoltaic power plants in desert locations, Case Stud. Therm. Eng. 47 (2023) 1–10.
45. [45] C. Carrillo, A.F. Obando Montaño, J. Cidrás, E. Díaz-Dorado, Review of power curve modelling for windturbines, Renew. Sustain. Energy Rev. 21 (2013) 572–581.
46. [46] A. Sedaghat, A. Hassanzadeh, J. Jamali, A. Mostafaeipour, W.H. Chen, Determination of rated wind speed for maximum annual energy production of variable speed wind turbines, Appl. Energy 205 (2017) 781–789.
47. [47] G.N. Güğül, G.D. Başbilen, D.K. Baker, Techno-Economic Analysis for Wind Energy Projects: A Comparative Study With Three Wind Turbines Based on Real-Site Data, Turkish J. Electr. Power Energy Syst. 3 (2023) 115–124.
48. [48] P. del R. Miguel Á. Rodríguez-López, Emilio Cerdá, Modeling Wind-Turbine Power Curves : Effects of Energy Generation, Energies (2020).
49. [49] S.H. Danook, K.J. Jassim, A.M. Hussein, The impact of humidity on performance of wind turbine, Case Stud. Therm. Eng. 14 (2019) 100456.
50. [50] M.H. El-Ahmar, A.H.M. El-Sayed, A.M. Hemeida, Evaluation of factors affecting wind turbine output power, 2017 19th Int. Middle-East Power Syst. Conf. MEPCON 2017 - Proc. 2018-February (2017) 1471–1476.