INTELLIGENT ARTIFICIAL NEURAL NETWORKS FOR LOW-LATENCY FAULT DETECTION AND PRECISE LOCALIZATION IN ELECTRICAL TRANSMISSION LINES: A CRITICAL REVIEW

Artificial intelligence has substantially improved the ability to detect, classify, and locate transmission-line faults from voltage, current, PMU, relay, and inspection data. The literature demonstrates that classical ANN, ANFIS, ELM, boosted trees, CNN, LSTM, TCN, transformer, GNN, transfer-learning, federated, explainable, and physics-informed approaches each address different parts of the problem. Reported simulation accuracy is often excellent, but field and hardware evidence shows that domain shift, topology, data quality, and complete latency remain decisive. The most credible future contribution is not a marginally higher random-split score. It is a unified, calibrated, physically constrained, topology-adaptive, low-computation framework validated on unseen events and measured from acquisition to trip decision. Such a framework can convert promising data-driven diagnosis into dependable protection support while retaining deterministic fallback and operator trust.

[1] V. N. Ogar, S. Hussain, and K. A. A. Gamage, “The use of artificial neural network for low latency of fault detection and localisation in transmission line,” Heliyon, vol. 9, no. 2, Art. no. e13376, 2023, doi: 10.1016/j.heliyon.2023.e13376. [2] K. Al Kharusi, A. El Haffar, and M. Mesbah, “Adaptive machine-learning-based transmission line fault detection and classification connected to inverter-based generators,” Energies, vol. 16, no. 15, Art. no. 5775, 2023, doi: 10.3390/en16155775. [3] L. Brito Palma, “Hybrid approach for detection and diagnosis of short-circuit faults in power transmission lines,” Energies, vol. 17, no. 9, Art. no. 2169, 2024, doi: 10.3390/en17092169. [4] A. Afonin and M. Chertkov, “Which neural network to choose for post-fault localization, dynamic state estimation, and optimal measurement placement in power systems?” Frontiers in Big Data, vol. 4, Art. no. 692493, 2021, doi: 10.3389/fdata.2021.692493. [5] A. S. Alhanaf, H. H. Balik, and M. Farsadi, “Intelligent fault detection and classification schemes for smart grids based on deep neural networks,” Energies, vol. 16, no. 22, Art. no. 7680, 2023, doi: 10.3390/en16227680. [6] Ö. Özdemir, R. Köker, and N. Pamuk, “Fault classification and precise fault location detection in 400 kV high-voltage power transmission lines using machine learning algorithms,” Processes, vol. 13, no. 2, Art. no. 527, 2025, doi: 10.3390/pr13020527. [7] M. S. Alam, M. R. Islam, R. Fan, M. S. A. Shazid, and A. S. Hasan, “Attention-guided multi-task learning for fault detection, classification, and localization in power transmission systems,” Energies, vol. 18, no. 24, Art. no. 6547, 2025, doi: 10.3390/en18246547. [8] M. Almutairi and W. Ko, “A hybrid deep learning and PINN approach for fault detection and classification in HVAC transmission systems,” Energies, vol. 18, no. 18, Art. no. 4796, 2025, doi: 10.3390/en18184796. [9] S. A. Rocha, T. G. de Mattos, and E. G. da Silveira, “A practical approach for fault location in transmission lines with series compensation using artificial neural networks: Results with field data,” Energies, vol. 18, no. 1, Art. no. 145, 2025, doi: 10.3390/en18010145. [10] M.-Y. Lee, Y.-S. Huang, C.-J. Chang, J.-Y. Yang, C.-W. Liu, and Y.-B. Lin, “Transmission line fault classification using Conformer convolution-augmented transformer model,” Applied Sciences, vol. 14, no. 10, Art. no. 4031, 2024, doi: 10.3390/app14104031.