One-Cycle Fault Classification and Faulted-Line Identification on the PROTECT-90 Dataset: An Initial Application Benchmark

2026-10-02

AI Summary

This study addresses the absence of application-level benchmarks with defined temporal and sensing assumptions in power system protection, which hinders evaluating the real-world utility of learning-based models. Building upon the PROTECT-90 dataset, this work proposes the first application benchmark for fault classification and line identification, employing a rigorous disjoint-split strategy to systematically evaluate 1D CNNs and MLPs across varying observation windows. Results demonstrate near-saturated accuracy under full observability. Notably, using current-only inputs maintains 100% line identification accuracy, whereas voltage-only inputs degrade performance to approximately 53%, with CPU inference requiring merely 0.528 ms. By quantifying the impact of sensor deficiencies on model precision, this research confirms that the fundamental bottleneck in such systems lies in measurement information rather than network architecture.

Abstract

Open electromagnetic-transient datasets are beginning to make reproducible learning-based protection studies possible, but the practical use of these datasets still requires application-level benchmarks that define timing, sensing, and validation assumptions. This paper presents an initial application benchmark on the recently released PROTECT-90 dataset for two protection-oriented tasks: fault-type classification and discrete faulted-line identification. A compact one-dimensional convolutional neural network (CNN) is evaluated using post-inception windows of 0.25, 0.5, 1, and 2 cycles under strict episode-wise splitting. A non-convolutional multilayer perceptron (MLP) is also trained as an architecture-control baseline. The results show that both tasks are nearly saturated under full observability, with one-cycle test accuracies of 99.84% for fault type and 100.00% for line identification. The main performance variation appears under reduced observability: current-only inputs preserve line identification accuracy at 100.00%, whereas voltage-only inputs reduce line identification accuracy to 53.09% with the CNN and 50.57% with the MLP. This indicates that the limiting factor is measurement information rather than neural architecture. Additional stratified checks show stable performance across topology states and fault-resistance bins, while CPU inference contributes only 0.528 ms to the one-cycle total decision time of 20.53 ms.

Problem

• Fault Classification
• Faulted-Line Identification
• Application Benchmark
• Electromagnetic Transients
• Power System Protection

Innovation

• Application Benchmark
• One-Cycle Fault Classification
• 1D-CNN
• Reduced Observability
• PROTECT-90 Dataset