Digital-twin and Physics-informed Neural Ensemble for Locating Asymmetrical Short-circuit Faults on 10 kV Overhead Feeders

Ramziddin Maxamatxodjayev Xasan o‘g‘li; Nurmatov Obid Yoqubboyevich

doi:doi:10.11648/j.epes.20261505.11

Research Article |

| Peer-Reviewed

Digital-twin and Physics-informed Neural Ensemble for Locating Asymmetrical Short-circuit Faults on 10 kV Overhead Feeders

Ramziddin Maxamatxodjayev Xasan o‘g‘li^*

, Nurmatov Obid Yoqubboyevich

Published in American Journal of Electrical Power and Energy Systems (Volume 15, Issue 5)

Received: 27 August 2026 Accepted: 4 September 2026 Published: 22 September 2026

Views: Downloads:

Download PDF

Share This Article

Twitter
Linked In
Facebook

Abstract

This paper proposes a digital-twin-assisted, physics-informed (feature-enriched) ensemble for locating asymmetrical short-circuit faults on a 10 kV radial overhead feeder using single-ended current and voltage measurements. The digital twin randomizes feeder sequence impedances, source impedance, loading, fault location, fault resistance, and measurement error to generate 6,000 operating scenarios. Observable fault cases are selected using an explicit current-based criterion, after which a data-only multilayer perceptron, a physics-informed multilayer perceptron, Random Forest, Gradient Boosting, and HistGradientBoosting are evaluated on an independent test subset. A weighted ensemble, 0.4 MLP + 0.6 HGB, achieves a mean absolute error of 0.962 km, an RMSE of 1.316 km, and an R² of 0.948, reducing MAE by approximately 25% relative to the data-only MLP. Error distributions are further analyzed with respect to fault resistance, measurement noise, observability, and uncertainty. The proposed model is not intended to replace deterministic protection; instead, it operates as a confidence-gated advisory layer that narrows the patrol segment and abstains under unreliable conditions. An IEC 61850-compatible deployment architecture, cybersecurity controls, a TRL 3-7 validation roadmap, and the remaining requirements for HIL, COMTRADE, and seasonal field validation are also presented.

Published in	American Journal of Electrical Power and Energy Systems (Volume 15, Issue 5)
DOI	10.11648/j.epes.20261505.11
Page(s)	103-114
Creative Commons	This is an Open Access article, distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium or format, provided the original work is properly cited.
Copyright	Copyright © The Author(s), 2026. Published by Science Publishing Group

Keywords

10 kV Overhead Feeder, Asymmetrical Short Circuit, Digital Twin, Physics-informed AI, Fault Location, Uncertainty, Relay Protection

References

[1]	Regional Electric Networks JSC (Hududiy elektr tarmoqlari AJ). Official data on 0.4–110 kV power networks and 10 kV lines, 2025.
[2]	IEC 60909-0: 2026. Short-circuit currents in three-phase AC systems — Part 0: Calculation of currents. IEC, 2026.
[3]	IEC 60909-3: 2009. Short-circuit currents in three-phase AC systems — Part 3. IEC, 2009.
[4]	IEC 60255-1: 2022. Measuring relays and protection equipment — Common requirements. IEC, 2022.
[5]	IEC 60255-151: 2009. Functional requirements for over/under current protection. IEC, 2009.
[6]	IEC TS 60255-216-1: 2025. Digital interface for protection functions using digital communication. IEC, 2025.
[7]	IEC 61850: 2026 SER. Communication networks and systems for power utility automation. IEC, 2026.
[8]	IEC TR 61850-90-21: 2025. Travelling Wave Fault Location. IEC, 2025.
[9]	IEC 61869-1: 2023. Instrument transformers — General requirements. IEC, 2023.
[10]	IEEE C37.112-2018. Inverse-Time Characteristics Equations for Overcurrent Relays. IEEE, 2018.
[11]	Martinez-Velasco J. A. et al. Survey on Methods for Detection, Classification and Location of Faults in Power Systems Using Artificial Intelligence. arXiv: 2507.10011, 2025.
[12]	Dong H. et al. A Method for Fault Localization in Distribution Networks with High Proportions of Distributed Generation Based on Graph Convolutional Networks. Energies, 2024, 17, 5758. https://doi.org/10.3390/en17225758
[13]	Liang L., Zhang H., Cao S., Zhao X., Li H., Chen Z. Fault location method for distribution networks based on multi-head graph attention networks. Frontiers in Energy Research, 2024, 12: 1395737.
[14]	Fan X. et al. Fault Location Method of Distribution Network Based on VGAE-GraphSAGE. Processes, 2024, 12, 2179.
[15]	Application of a Centroid Frequency-Based Back Propagation Neural Network Fault Location Method. Electronics, 2024, 13, 1491.
[16]	Fault Line Selection Method for Power Distribution Network Based on Graph Transformation and ResNet50 Model. Information, 2024, 15, 375.
[17]	Thurner L., Braun M. Vectorized Calculation of Short Circuit Currents Considering Distributed Generation. IEEE ISGT Europe, 2018. https://doi.org/10.1109/ISGTEurope.2018.8571529
[18]	Majidi M., Etezadi-Amoli M. A New Fault Location Technique in Smart Distribution Networks. IEEE Transactions on Power Delivery, 2018, 33(3), 1358–1368.
[19]	Nayak S. et al. Fault Detection in Electrical Distribution System using Autoencoders. arXiv: 2602.14939, 2026.
[20]	Jeha J., Arunachalam K. Advanced fault detection and classification in high voltage transmission lines using deep learning integration. Evolving Systems, 2026, 17: 12. https://doi.org/10.1007/s12530-025-09777-8

Cite This Article

Plain Text BibTeX RIS

APA Style

o‘g‘li, R. M. X., Yoqubboyevich, N. O. (2026). Digital-twin and Physics-informed Neural Ensemble for Locating Asymmetrical Short-circuit Faults on 10 kV Overhead Feeders. American Journal of Electrical Power and Energy Systems, 15(5), 103-114. https://doi.org/10.11648/j.epes.20261505.11

Copy | Download

ACS Style

o‘g‘li, R. M. X.; Yoqubboyevich, N. O. Digital-twin and Physics-informed Neural Ensemble for Locating Asymmetrical Short-circuit Faults on 10 kV Overhead Feeders. Am. J. Electr. Power Energy Syst. 2026, 15(5), 103-114. doi: 10.11648/j.epes.20261505.11

Copy | Download

AMA Style

o‘g‘li RMX, Yoqubboyevich NO. Digital-twin and Physics-informed Neural Ensemble for Locating Asymmetrical Short-circuit Faults on 10 kV Overhead Feeders. Am J Electr Power Energy Syst. 2026;15(5):103-114. doi: 10.11648/j.epes.20261505.11

Copy | Download

@article{10.11648/j.epes.20261505.11,
  author = {Ramziddin Maxamatxodjayev Xasan o‘g‘li and Nurmatov Obid Yoqubboyevich},
  title = {Digital-twin and Physics-informed Neural Ensemble for Locating Asymmetrical Short-circuit Faults on 10 kV Overhead Feeders},
  journal = {American Journal of Electrical Power and Energy Systems},
  volume = {15},
  number = {5},
  pages = {103-114},
  doi = {10.11648/j.epes.20261505.11},
  url = {https://doi.org/10.11648/j.epes.20261505.11},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.epes.20261505.11},
  abstract = {This paper proposes a digital-twin-assisted, physics-informed (feature-enriched) ensemble for locating asymmetrical short-circuit faults on a 10 kV radial overhead feeder using single-ended current and voltage measurements. The digital twin randomizes feeder sequence impedances, source impedance, loading, fault location, fault resistance, and measurement error to generate 6,000 operating scenarios. Observable fault cases are selected using an explicit current-based criterion, after which a data-only multilayer perceptron, a physics-informed multilayer perceptron, Random Forest, Gradient Boosting, and HistGradientBoosting are evaluated on an independent test subset. A weighted ensemble, 0.4 MLP + 0.6 HGB, achieves a mean absolute error of 0.962 km, an RMSE of 1.316 km, and an R2 of 0.948, reducing MAE by approximately 25% relative to the data-only MLP. Error distributions are further analyzed with respect to fault resistance, measurement noise, observability, and uncertainty. The proposed model is not intended to replace deterministic protection; instead, it operates as a confidence-gated advisory layer that narrows the patrol segment and abstains under unreliable conditions. An IEC 61850-compatible deployment architecture, cybersecurity controls, a TRL 3-7 validation roadmap, and the remaining requirements for HIL, COMTRADE, and seasonal field validation are also presented.},
 year = {2026}
}

Copy | Download

TY  - JOUR
T1  - Digital-twin and Physics-informed Neural Ensemble for Locating Asymmetrical Short-circuit Faults on 10 kV Overhead Feeders
AU  - Ramziddin Maxamatxodjayev Xasan o‘g‘li
AU  - Nurmatov Obid Yoqubboyevich
Y1  - 2026/09/22
PY  - 2026
N1  - https://doi.org/10.11648/j.epes.20261505.11
DO  - 10.11648/j.epes.20261505.11
T2  - American Journal of Electrical Power and Energy Systems
JF  - American Journal of Electrical Power and Energy Systems
JO  - American Journal of Electrical Power and Energy Systems
SP  - 103
EP  - 114
PB  - Science Publishing Group
SN  - 2326-9200
UR  - https://doi.org/10.11648/j.epes.20261505.11
AB  - This paper proposes a digital-twin-assisted, physics-informed (feature-enriched) ensemble for locating asymmetrical short-circuit faults on a 10 kV radial overhead feeder using single-ended current and voltage measurements. The digital twin randomizes feeder sequence impedances, source impedance, loading, fault location, fault resistance, and measurement error to generate 6,000 operating scenarios. Observable fault cases are selected using an explicit current-based criterion, after which a data-only multilayer perceptron, a physics-informed multilayer perceptron, Random Forest, Gradient Boosting, and HistGradientBoosting are evaluated on an independent test subset. A weighted ensemble, 0.4 MLP + 0.6 HGB, achieves a mean absolute error of 0.962 km, an RMSE of 1.316 km, and an R2 of 0.948, reducing MAE by approximately 25% relative to the data-only MLP. Error distributions are further analyzed with respect to fault resistance, measurement noise, observability, and uncertainty. The proposed model is not intended to replace deterministic protection; instead, it operates as a confidence-gated advisory layer that narrows the patrol segment and abstains under unreliable conditions. An IEC 61850-compatible deployment architecture, cybersecurity controls, a TRL 3-7 validation roadmap, and the remaining requirements for HIL, COMTRADE, and seasonal field validation are also presented.
VL  - 15
IS  - 5
ER  -

Copy | Download

Author Information

Ramziddin Maxamatxodjayev Xasan o‘g‘li

The Department of Mechanics, Tashkent State Technical University, Tashkent, Uzbekistan

Contact Email

http://orcid.org/0009-0004-0751-4175
Nurmatov Obid Yoqubboyevich

The Department of Mechanics, Tashkent State Technical University, Tashkent, Uzbekistan

Contact Email

Download PDF

Submit an Article

Sections

Plain Text BibTeX RIS

APA Style

o‘g‘li, R. M. X., Yoqubboyevich, N. O. (2026). Digital-twin and Physics-informed Neural Ensemble for Locating Asymmetrical Short-circuit Faults on 10 kV Overhead Feeders. American Journal of Electrical Power and Energy Systems, 15(5), 103-114. https://doi.org/10.11648/j.epes.20261505.11

Copy | Download

ACS Style

o‘g‘li, R. M. X.; Yoqubboyevich, N. O. Digital-twin and Physics-informed Neural Ensemble for Locating Asymmetrical Short-circuit Faults on 10 kV Overhead Feeders. Am. J. Electr. Power Energy Syst. 2026, 15(5), 103-114. doi: 10.11648/j.epes.20261505.11

Copy | Download

AMA Style

o‘g‘li RMX, Yoqubboyevich NO. Digital-twin and Physics-informed Neural Ensemble for Locating Asymmetrical Short-circuit Faults on 10 kV Overhead Feeders. Am J Electr Power Energy Syst. 2026;15(5):103-114. doi: 10.11648/j.epes.20261505.11

Copy | Download

@article{10.11648/j.epes.20261505.11,
  author = {Ramziddin Maxamatxodjayev Xasan o‘g‘li and Nurmatov Obid Yoqubboyevich},
  title = {Digital-twin and Physics-informed Neural Ensemble for Locating Asymmetrical Short-circuit Faults on 10 kV Overhead Feeders},
  journal = {American Journal of Electrical Power and Energy Systems},
  volume = {15},
  number = {5},
  pages = {103-114},
  doi = {10.11648/j.epes.20261505.11},
  url = {https://doi.org/10.11648/j.epes.20261505.11},
  eprint = {https://article.sciencepublishinggroup.com/pdf/10.11648.j.epes.20261505.11},
  abstract = {This paper proposes a digital-twin-assisted, physics-informed (feature-enriched) ensemble for locating asymmetrical short-circuit faults on a 10 kV radial overhead feeder using single-ended current and voltage measurements. The digital twin randomizes feeder sequence impedances, source impedance, loading, fault location, fault resistance, and measurement error to generate 6,000 operating scenarios. Observable fault cases are selected using an explicit current-based criterion, after which a data-only multilayer perceptron, a physics-informed multilayer perceptron, Random Forest, Gradient Boosting, and HistGradientBoosting are evaluated on an independent test subset. A weighted ensemble, 0.4 MLP + 0.6 HGB, achieves a mean absolute error of 0.962 km, an RMSE of 1.316 km, and an R2 of 0.948, reducing MAE by approximately 25% relative to the data-only MLP. Error distributions are further analyzed with respect to fault resistance, measurement noise, observability, and uncertainty. The proposed model is not intended to replace deterministic protection; instead, it operates as a confidence-gated advisory layer that narrows the patrol segment and abstains under unreliable conditions. An IEC 61850-compatible deployment architecture, cybersecurity controls, a TRL 3-7 validation roadmap, and the remaining requirements for HIL, COMTRADE, and seasonal field validation are also presented.},
 year = {2026}
}

Copy | Download

TY  - JOUR
T1  - Digital-twin and Physics-informed Neural Ensemble for Locating Asymmetrical Short-circuit Faults on 10 kV Overhead Feeders
AU  - Ramziddin Maxamatxodjayev Xasan o‘g‘li
AU  - Nurmatov Obid Yoqubboyevich
Y1  - 2026/09/22
PY  - 2026
N1  - https://doi.org/10.11648/j.epes.20261505.11
DO  - 10.11648/j.epes.20261505.11
T2  - American Journal of Electrical Power and Energy Systems
JF  - American Journal of Electrical Power and Energy Systems
JO  - American Journal of Electrical Power and Energy Systems
SP  - 103
EP  - 114
PB  - Science Publishing Group
SN  - 2326-9200
UR  - https://doi.org/10.11648/j.epes.20261505.11
AB  - This paper proposes a digital-twin-assisted, physics-informed (feature-enriched) ensemble for locating asymmetrical short-circuit faults on a 10 kV radial overhead feeder using single-ended current and voltage measurements. The digital twin randomizes feeder sequence impedances, source impedance, loading, fault location, fault resistance, and measurement error to generate 6,000 operating scenarios. Observable fault cases are selected using an explicit current-based criterion, after which a data-only multilayer perceptron, a physics-informed multilayer perceptron, Random Forest, Gradient Boosting, and HistGradientBoosting are evaluated on an independent test subset. A weighted ensemble, 0.4 MLP + 0.6 HGB, achieves a mean absolute error of 0.962 km, an RMSE of 1.316 km, and an R2 of 0.948, reducing MAE by approximately 25% relative to the data-only MLP. Error distributions are further analyzed with respect to fault resistance, measurement noise, observability, and uncertainty. The proposed model is not intended to replace deterministic protection; instead, it operates as a confidence-gated advisory layer that narrows the patrol segment and abstains under unreliable conditions. An IEC 61850-compatible deployment architecture, cybersecurity controls, a TRL 3-7 validation roadmap, and the remaining requirements for HIL, COMTRADE, and seasonal field validation are also presented.
VL  - 15
IS  - 5
ER  -

Copy | Download