A Hybrid Blockchain and Machine Learning Framework for Loss-Aware Optimization in Distributed Photovoltaic Systems

Abstract

This study proposes a hybrid framework integrating machine learning and blockchain technologies for loss-aware optimization in distributed photovoltaic (PV) systems. The framework combines data-driven loss prediction, decentralized data validation, and optimization-based energy management to improve operational efficiency under variable generation and demand conditions. An LSTM-based model is employed to estimate system losses, while a blockchain layer with smart contracts provides transparent and tamper-resistant execution of operational decisions. The predicted losses are incorporated into the optimization process to determine more efficient energy-routing and grid-interaction strategies. Simulation-based evaluation demonstrates that the proposed framework outperforms the baseline and machine-learning-only configurations. In the investigated case, the energy-loss rate decreases from 14.2% in the baseline scenario to 6.3% with the proposed framework, corresponding to a 55.6% relative reduction. The results also indicate improvements in operational cost and overall system efficiency. These findings demonstrate the potential of combining predictive intelligence with decentralized validation for efficient and secure management of distributed PV systems.

Keywords

• Photovoltaic systems
• energy loss optimization
• machine learning
• blockchain technology
• smart grids

Dağıtılmış Fotovoltaik Sistemlerde Kayıp Duyarlı Optimizasyon için Hibrit Blok Zinciri ve Makine Öğrenimi Çerçevesi

Abstract

Bu çalışma, dağıtılmış fotovoltaik (PV) sistemlerde kayıp odaklı optimizasyon için makine öğrenimi ve blok zinciri teknolojilerini entegre eden hibrit bir çerçeve önermektedir. Çerçeve, değişken üretim ve talep koşulları altında operasyonel verimliliği artırmak için veri odaklı kayıp tahmini, merkezi olmayan veri doğrulaması ve optimizasyon tabanlı enerji yönetimini bir araya getirmektedir. Sistem kayıplarını tahmin etmek için LSTM tabanlı bir model kullanılırken, akıllı sözleşmelere sahip bir blok zinciri katmanı, operasyonel kararların şeffaf ve kurcalamaya karşı dayanıklı bir şekilde yürütülmesini sağlar. Tahmin edilen kayıplar, daha verimli enerji yönlendirme ve şebeke etkileşim stratejilerini belirlemek için optimizasyon sürecine dahil edilir. Simülasyon tabanlı değerlendirme, önerilen çerçevenin temel ve yalnızca makine öğrenimi tabanlı yapılandırmalardan daha iyi performans gösterdiğini ortaya koymaktadır. İncelenen durumda, enerji kaybı oranı temel senaryoda %14,2'den önerilen çerçeve ile %6,3'e düşmekte olup, bu da %55,6'lık bir göreceli azalmaya karşılık gelmektedir. Sonuçlar ayrıca operasyonel maliyetlerde ve genel sistem verimliliğinde iyileşmeler olduğunu göstermektedir. Bu bulgular, dağıtılmış PV sistemlerinin verimli ve güvenli yönetimi için tahmine dayalı zekayı merkezi olmayan doğrulama ile birleştirmenin potansiyelini göstermektedir.

Keywords

• Fotovoltaik sistemler
• enerji kayıpları optimizasyonu
• makine öğrenimi
• blok zinciri teknolojisi
• akıllı şebekeler

References

1. Zhu Y, Xu X, Yan Z, Lu J. Data acquisition, power forecasting and coordinated dispatch of power systems with distributed PV power generation. The Electricity Journal 2022; 35(5): 107133.
2. Jiang Y, Li X, Yao J, Wan X, Zhang J, Dai Y. Design and performance simulation of a distributed aerobic composting system assisted by solar PV/T heat pump. Renewable Energy 2022; 196: 547-559.
3. Nazir M, Tyagi A, Tyagi VV, Kumar K, Krishan R. A review on reactive power capability of distributed solar PV inverter in distribution systems. International Journal of Ecology and Sustainable Development 2022; 13(2).
4. Baran E. A multi-criteria decision-making approach integrated with machine learning for energy resource supply. Systems 2026; 14(2): 200.
5. Chen X, Xu F, He G, Li Z, Wang F, Li K, Catalao JPS. Decoupling based monthly net electricity consumption prediction model considering high penetration of distributed solar PV systems. Sustainable Energy, Grids and Networks 2022; 32: 100870.
6. KorkusuzPolat T, Baran E. A blockchain-based quality 4.0 application for warehouse management system. Applied Systems 2024; 14(23): 10950.
7. Le Y, Jie Y, Zhongyao Z. Blockchain-based distributed PV local consumption with multi-agent nash bargaining. IFAC-PapersOnLine 2025; 59(34): 54-59.
8. Kavin R, Jayakumar J. Energy management system for distributed energy resources using blockchain technology. Recent Patents on Engineering 2025; 19(1).

9. Baran E. Deep-learning based detection and forecasting of performance losses in solar PV systems using multi-sensor data. Applied Sciences 2026; 16(4): 1709.
10. Lan Y, Illindala M. Resilient distributed load frequency control for solar PV-dominated interconnected power systems against PMU denial-of-service attacks. International Journal of Electrical Power & Energy Systems 2025; 172: 111188.
11. Ni J, Chen Y, Goudarzi A, Wang T, Yang L, Mei S. A dual time-scale optimal dispatch algorithm for PV systems: integrating centralized optimal power dispatch with distributed power deviation absorption in DC smart grids. Cleaner Engineering and Technology 2025; 26: 100961.
12. Wang X, An T, Sui S, Huang F, Zhu Q, Song J, Lu Z, Zhu J. Dynamic aggregation-based federated learning for fault diagnosis of distributed PV systems. Electric Power Systems Research 2025; 246: 111648.
13. Shoaib MR, Emara HM, Zhao J, Ahvanooey MT, Nabil E. Advanced deep learning approaches for fault detection in solar PV systems: a comparative study of SPDA and AIFD-So1DL. Applied Soft Computing 2025; 184: 113592.
14. Zhang B, Cao D, Hu W, Ghias AMYM, Chen Z. Physics-informed multi-agent deep reinforcement learning enabled distributed voltage control for active distribution network using PV inverters. International Journal of Electrical Power & Energy Systems 2025; 155: 109641.
15. Francisti J, Fodor K, Balogh Z, Magdin M. Predictive modeling and anomaly detection in solar PV inverters using machine learning. Results in Engineering 2025; 28: 108348.
16. Irkam A, Hannan MA, Rahman SA, Roslan MF, Ker PJ, Wong RTK, Jang G. Evaluation of critical usage duration for PV/BES and PV/BES/H2 systems with machine learning models. Journal of Energy Storage 2025; 136: 118414.
17. Al-Dahidi S, Alrbai M, Rinchi B, Al-Gussain L, Ayadi O, Alahmer A. A tiered NARX model for forecasting da-ahead energy production in distributed solar PV systems. Cleaner Engineering and Technology 2024; 23: 100831.
18. Renold AP, Sinha N, Gao XZ. Hybrid machine learning approach for improved short-term PV power forecasting accuracy. Results in Engineering 2025; 28: 107374.
19. Maji S, Kayal P. A simplified multi-objective planning approach for allocation of distributed PV generators in unbalanced power distribution systems. Renewable Energy Focus 2024; 48: 100541.
20. Baran E, KorkusuzPolat T. Classification of industry 4.0 for total quality management: a review. Sustainability 2022; 14(6): 3329.
21. Zhao M, Liao L, Liang P, Li M, Dong X, Yang Y, Wang H, Zhang Z. Energy blockchain in smart communities: towards affordable clean energy supply for the built environment. Energy Engineering 2024; 121(8): 2313-2330.
22. Li Y, Tao Q, Gong Y. Digital twin simulation for integration of blockchain and internet of things for optimal smart management of PV-based connected microgrids. Solar Energy 2023; 251: 306-314.
23. Wang H, Feng T, Xiong W. How can the dynamic game be integrated into blockchain-based distributed energy resources multi-agent transactions for decision-making. Energy 2022; 254: 124318.
24. Chadly A, Hasani HR, Moawad K, Salah K, Omar M, Mayyas A. A blockchain-based solution for the traceability of rare earth metals used in thin-film photovoltaics. Journal of Cleaner Production 2023; 428: 139399.
25. Kashyap PK, Dohare U, Kumar M, Kumar S. Blockchain and quantum machine learning driven energy trading for electric vehicles. Ad Hoc Networks 2024; 165: 103632.
26. Faheem M, Al-Khasawneh, MA. Multilayer cyberattacks identification and classification using machine learning in internet of blockchain (IoBC)-based energy networks. Data in Brief 2024; 54: 110461.
27. Minh NO, Al-Khiem NT, Giang VH. Fault classification and localization in power transmission line based on machine learning and combined CNN-LSTM models. Energy Reports 2024; 12: 5610-5622.