Improving Electrical Fault Detection Using Multiple Classifier Systems DOI Creative Commons

José Gerardo Beserra de Oliveira,

Dioéliton Passos, Davi Carvalho

и другие.

Energies, Год журнала: 2024, Номер 17(22), С. 5787 - 5787

Опубликована: Ноя. 20, 2024

Machine Learning-based fault detection approaches in energy systems have gained prominence for their superior performance. These automated can assist operators by highlighting anomalies and faults, providing a robust framework improving Situation Awareness. However, existing predominantly rely on monolithic models, which struggle with adapting to changing data, handling imbalanced datasets, capturing patterns noisy environments. To overcome these challenges, this study explores the potential of Multiple Classifier System (MCS) approaches. The results demonstrate that ensemble methods generally outperform single dynamic like META-DES showing remarkable resilience noise. findings highlight importance model diversity strategies classification accuracy under real-world, conditions. This research emphasizes MCS techniques as solution enhancing reliability systems.

Язык: Английский

Computational Methods, Artificial Intelligence, Modeling, and Simulation Applications in Green Hydrogen Production Through Water Electrolysis: A Review DOI Creative Commons
Ahmed Y. Shash,

Noha M. Abdeltawab,

Doaa Hassan

и другие.

Hydrogen, Год журнала: 2025, Номер 6(2), С. 21 - 21

Опубликована: Март 25, 2025

Green hydrogen production is emerging as a crucial component in global decarbonization efforts. This review focuses on the role of computational approaches and artificial intelligence (AI) optimizing green technologies. Key to improving electrolyzer efficiency scalability include fluid dynamics (CFD), thermodynamic modeling, machine learning (ML). As an instance, CFD has achieved over 95% accuracy estimating flow distribution polarization curves, but AI-driven optimization can lower operational expenses by up 24%. Proton exchange membrane electrolyzers achieve efficiencies 65–82% at 70–90 °C, solid oxide reach 90% temperatures ranging from 650 1000 °C. According studies, combining renewable energy with reduces emissions improves grid reliability, curtailment rates less than 1% for biomass-driven systems. integration ensures long-term transition while also addressing security environmental concerns.

Язык: Английский

Процитировано

0

Edge Artificial Intelligence for Electrical Anomaly Detection Based on Process-In-Memory Chip DOI Open Access
Jian-Ming Jin, Xiang Qiu, Cimang Lu

и другие.

Electronics, Год журнала: 2024, Номер 13(21), С. 4255 - 4255

Опубликована: Окт. 30, 2024

Neural-networks (NNs) for the current feature analysis bring novel electrical safety functions in smart circuit breakers (CBs), especially preventing fire hazard from electric vehicle/bike battery charging. In this work, edge artificial intelligence (AI) solutions anomaly detection were designed and demonstrated based on process-in-memory (PIM) AI chip. The ultra-low power high-performance character of PIM chips enable solution to embed limited space inside breaker detect improper charging at millisecond latency.

Язык: Английский

Процитировано

0

Improving Electrical Fault Detection Using Multiple Classifier Systems DOI Creative Commons

José Gerardo Beserra de Oliveira,

Dioéliton Passos, Davi Carvalho

и другие.

Energies, Год журнала: 2024, Номер 17(22), С. 5787 - 5787

Опубликована: Ноя. 20, 2024

Machine Learning-based fault detection approaches in energy systems have gained prominence for their superior performance. These automated can assist operators by highlighting anomalies and faults, providing a robust framework improving Situation Awareness. However, existing predominantly rely on monolithic models, which struggle with adapting to changing data, handling imbalanced datasets, capturing patterns noisy environments. To overcome these challenges, this study explores the potential of Multiple Classifier System (MCS) approaches. The results demonstrate that ensemble methods generally outperform single dynamic like META-DES showing remarkable resilience noise. findings highlight importance model diversity strategies classification accuracy under real-world, conditions. This research emphasizes MCS techniques as solution enhancing reliability systems.

Язык: Английский

Процитировано

0