Real-time Evaluation of Relay Protection System Status for Smart Grid: A Fusion Model of Digital Twin and Deep Transfer Learning
DOI:
https://doi.org/10.13052/spee1048-5236.45212Keywords:
Relay protection, digital twin, deep transfer learning, sparse autoencoder, real-time state evaluation, intelligent setting managementAbstract
To address issues in traditional relay protection system state evaluation, such as insufficient training samples, lagging results, and manual setting management, this study proposes a real-time state evaluation model integrating digital twin and deep transfer learning. A high-fidelity digital twin system is constructed to establish bidirectional mapping and dynamic updates between the physical system and virtual twin. A sparse stacked autoencoder extracts discriminative features, and an online adaptation strategy based on deep transfer learning enables continuous self-optimization with streaming data. Experimental results show an overall accuracy of 98.2%, weighted F1-score of 0.978, and average evaluation delay reduced to 1.5 minutes. The intelligent setting management platform improves verification and download efficiency by 60% and 40% respectively, with error rate decreasing from 10.2% to 0.22%. The framework enables minute-level real-time evaluation, predictive maintenance, and closed-loop operation, providing a reliable approach for building resilient and smart grids.
Downloads
References
Qin B, Pan H, Dai Y, Si X, Huang X, Yuen C. Machine and deep learning for digital twin networks: A survey[J]. IEEE Internet of Things Journal, 2024, 11(21): 34694–34716. DOI:10.1109/JIOT.2024.3416733.
Yao Z, Li D, Li Z, Zhou P, Li L. Relay protection mirror operation technology based on digital twin[J]. Protection and Control of Modern Power Systems, 2023, 8(4): 1–14. DOI:10.1186/s41601-023-00328-4.
Khan M M S, Giraldo J, Parvania M. Real-time cyber attack localization in distribution systems using digital twin reference model[J]. Ieee transactions on power delivery, 2023, 38(5): 3238–3249. DOI:10.1109/TPWRD.2023.3296312.
Mani D, Harispuru C, Wetterstrand N, Bonetti A, Thota S C R. Revolutionizing power system protection: closed-loop testing with digital twin technology[C]//IET Conference Proceedings CP916. Stevenage, UK: The Institution of Engineering and Technology, 2025, 2025(5): 36–40. DOI:10.1049/icp.2025.1043.
Najar A, Kazemi Karegar H, Esmaeilbeigi S. Multi-agent protection scheme for microgrid using deep learning[J]. IET Renewable Power Generation, 2024, 18(4): 663–678. DOI:10.1049/rpg2.12929.
Sultana A, Bardalai A, Sarma K K. Salp swarm-artificial neural network based cyber-attack detection in smart grid[J]. Neural Processing Letters, 2022, 54(4): 2861–2883. DOI:10.1007/s11063-022-10743-7.
Mashal I. Smart grid reliability evaluation and assessment[J]. Kybernetes, 2023, 52(9): 3261–3291. DOI:10.1108/K-12-2020-0910.
Li F, Zhuyuan L. A summary of relay protection-based simulation for Dynamic Performance and Reliability Assessment[J]. International Journal for Applied Information Management, 2023, 3(1): 11–23. DOI:10.47738/ijaim.v3i1.46.
Hu Q, Han R, Quan X, Wu Z, Tang C, Li W. Grid-forming inverter enabled virtual power plants with inertia support capability[J]. IEEE Transactions on Smart Grid, 2022, 13(5): 4134–4143. DOI:10.1109/TSG.2022.3141414.
Li Y, Yu C, Shahidehpour M, Yang T, Zeng Z, Chai T. Deep reinforcement learning for smart grid operations: Algorithms, applications, and prospects[J]. Proceedings of the IEEE, 2023, 111(9): 1055–1096. DOI:10.1109/JPROC.2023.3303358.
Noman M, Ullah I, Khan M A, Qazi A, Farooq W, Saqr A, Elsheikh A. Analysis of overcurrent protective relaying as minimum adopted fault protection for small-scale hydropower plants[J]. International Journal of Environmental Science and Technology, 2024, 21(4): 4457–4470. DOI:10.1007/s13762-023-05284-y.
Ataee-Kachoee A H, Hashemi-Dezaki H, Ketabi A. Optimal adaptive protection of smart grids using high-set relays and smart selection of relay tripping characteristics considering different network configurations and operation modes[J]. IET generation, transmission & distribution, 2022, 16(24): 5084–5116. DOI:10.1049/gtd2.12659.
Shittu M A, Shittu H A, Adeleke O J, Adeleke O J. Digital Twin Modeling for Real Time Monitoring and Fault Detection in Smart Substations[J]. International Journal of Industrial Engineering, 2023, 14(2): 25–44. DOI:10.34218/IJIERD_14_02_003.
Grasmair M, Haltmeier M, Scherzer O. Sparse regularization with lq penalty term[J]. Inverse Problems, 2008, 24(5): 055020. DOI:10.1088/0266-5611/24/5/055020.
Tiwari R, Singh R K, Choudhary N K. Coordination of dual setting overcurrent relays in microgrid with optimally determined relay characteristics for dual operating modes[J]. Protection and Control of Modern Power Systems, 2022, 7(1): 1–18. DOI:10.1186/s41601-022-00226-1.
Ashraf S, Evkay I, Selamogullari U S, Baysal M, Hasan O. Performance analysis of the dual-setting directional overcurrent relays-based protection considering the impact of curve types and fault location[J]. Electric Power Components and Systems, 2023, 51(7): 706–723. DOI:10.1080/15325008.2023.2182840.
Srivastava A, Liu C C, Stefanov A, Basumallik S, Hussain M M, Somda B. Digital twins serving cybersecurity: More than a model: Cybersecurity as a future benefit of digital twins 2[J]. IEEE Power and Energy Magazine, 2024, 22(1): 61–71. DOI:10.1109/MPE.2023.3325196.
Arraño-Vargas F, Konstantinou G. Modular design and real-time simulators toward power system digital twins implementation[J]. IEEE Transactions on Industrial Informatics, 2022, 19(1): 52–61. DOI:10.1109/TII.2022.3178713.
Vikram Raju G, Srikanth N V. Single fuzzy inference based fault detection and classification protection scheme for different types of short circuit faults on double circuit transmission lines[J]. International Journal of Modelling and Simulation, 2025, 45(1): 218–236. DOI:10.1080/02286203.2023.2193912.
Önder M, Dogan M U, Polat K. Classification of smart grid stability prediction using cascade machine learning methods and the internet of things in smart grid[J]. Neural Computing and Applications, 2023, 35(24): 17851–17869. DOI:10.1007/s00521-023-08605-x.
Mendoza M A T, Segundo-Ramírez J, Gurrola A E, Visairo-Cruz N, Guitierrez C A N, Torres U. Digital twin adaptive remedial action scheme for preventing voltage collapse[J]. IEEE Journal of Emerging and Selected Topics in Industrial Electronics, 2024, 6(2): 523–535. DOI:10.1109/JESTIE.2024.3465429.
https://aeeeuropeenergy.com/
