arXiv:2510. 00831v2 Announce Type: replace Abstract: The increasing complexity of modern power systems, driven by the integration of inverter-based and distributed energy resources, challenges the reliability of conventional protection schemes and motivates the use of machine learning for protection tasks.
By Julian Oelhaf, Georg Kordowich, Changhun Kim, Paula Andrea P\'erez-Toro, Christian Bergler, Andreas Maier, Johann J\"ager, Siming Bayer
The paper introduces a hybrid two‑stage machine learning pipeline for fault detection and classification in high‑voltage transmission networks. Stage 1 uses an Isolation Forest anomaly detector combined with an optional supervised binary detector, while Stage 2 applies a Random Forest multiclass classifier only to samples flagged by Stage 1. Feature engineering maps six raw channels to eighteen features, including zero‑sequence symmetrical components, achieving end‑to‑end accuracies of 95.8 % on the TLFaultDataset and 97.25 % on an independent single‑point dataset, surpassing federated benchmarks without GPU or federated infrastructure.
By Sahil Manikshete, Atharva Gujarathi, Thanh Long Vu, Akhtar Hussain, Van-Hai Bui
This study evaluates large language models (LLMs) for predicting weather‑related forced outage risk in a distribution grid using a zero‑shot approach without labeled training data. The task is framed as binary severity classification over 3h, 6h, and 12h horizons, leveraging six years of outage records and high‑resolution weather data from central Texas. Four zero‑shot LLMs are compared to two supervised classifiers under two input settings—current weather observations and forecast data—showing that supervised models lead on macro‑F1 and precision, while newer LLMs achieve competitive scores and offer complementary strengths in reasoning and geographic scalability.
By Christos Petridis, Zoran Obradovic, Mladen Kezunovic
The paper presents a machine‑learning framework for classifying power‑system contingencies into safe, moderate, or severe categories. Using Newton‑Raphson load flow data, the study applies SMOTE, PCA, and classifiers (KNN, Random Forest, SVM) to IEEE‑14 and IEEE‑30 bus systems, evaluating performance with precision, recall, and F1 score. Random Forest achieved the highest F1 scores, while PCA improved overall performance more than SMOTE, which boosted recall at the cost of some false positives.
By Joshua Salako, Folajimi Osikomaiya, Olakorede Olamiju
The paper introduces learned control variates (CVs) to estimate rare grid‑outage probabilities at electric vehicle charging stations more efficiently. By training a classifier on inexpensive features such as temperature and load forecasts, the CV reduces variance in Monte Carlo simulations, with performance tied to the classifier’s recall and false‑positive rate. Experiments on a feeder‑overload simulator and a real Surabaya network show that even a weak learner can cut variance by a quarter, while a boosted‑tree learner achieves about 60% reduction, leading to a one‑third decrease in city‑wide blackout days compared to standard Monte Carlo.
By Sulyman Iyanda, Mansur Arief
arXiv:2609.16744v1 Announce Type: new
Abstract: The integration of renewable energy sources into the electrical grid introduces complex challenges in fault detection and coordination of grid recovery...
By Julian Oelhaf, Georg Kordowich, Paula Andrea P\'erez-Toro, Tom\'as Arias-Vergara, Andreas Maier, Johann J\"ager, Siming Bayer