Iskanje po katalogu digitalne knjižnice

Opis: The paper investigates the application of machine learning (ML) for voltage sag source localization (VSSL) in electrical power systems. To overcome feature-selection challenges for traditional ML methods and provide more meaningful sequential features for deep learning methods, the paper proposes three time-sample-based feature forms, and evaluates an existing feature form. The effectiveness of these feature forms is assessed using k-means clustering with k = 2 referred to as downstream and upstream classes, according to the direction of voltage sag origins. Through extensive voltage sag simulations, including noises in a regional electrical power network, k-means identifies a sequence involving the multiplication of positive-sequence current magnitude with the sine of its angle as the most prominent feature form. The study develops further traditional ML methods such as decision trees (DT), support vector machine (SVM), random forest (RF), k-nearest neighbor (KNN), an ensemble learning (EL), and a designed one-dimensional convolutional neural network (1D-CNN). The results found that the combination of 1D-CNN or SVM with the most prominent feature achieved the highest accuracies of 99.37% and 99.13%, respectively, with acceptable/fast prediction times, enhancing VSSL. The exceptional performance of the CNN was also approved by field measurements in a real power network. However, selecting the best ML methods for deployment requires a trade-off between accuracy and real-time implementation requirements. The research findings benefit network operators, large factory owners, and renewable energy park producers. They enable preventive maintenance, reduce equipment downtime/damage in industry and electrical power systems, mitigate financial losses, and facilitate the assignment of power-quality penalties to responsible parties.
Ključne besede: voltage sag (dip), source localization, supervised and unsupervised learning, convolutional neural network, time-sample-based features
Objavljeno v DKUM: 23.08.2024; Ogledov: 65; Prenosov: 14
Celotno besedilo (15,94 MB)
Gradivo ima več datotek! Več...

Opis: This paper presents a comprehensive study of winter temperatures in Norway and northern Sweden, covering a period of 50 to 70 years. The analysis utilizes Singular Spectrum Analysis (SSA) to investigate temperature trends at six selected locations. The results demonstrate an overall long-term rise in temperatures, which can be attributed to global warming. However, when investigating variations in highest, lowest, and average temperatures for December, January, and February, 50% of the cases exhibit a significant decrease in recent years, indicating colder winters, especially in December. The study also explores the variations in Atlantic Meridional Overturning Circulation (AMOC) variations as a crucial climate factor over the last 15 years, estimating a possible 20% decrease/slowdown within the first half of the 21st century. Subsequently, the study investigates potential similarities between winter AMOC and winter temperatures in the mid to high latitudes over the chosen locations. Additionally, the study examines another important climatic index, the North Atlantic Oscillation (NAO), and explores possible similarities between the winter NAO index and winter temperatures. The findings reveal a moderate observed lagged correlation for AMOC-smoothed temperatures, particularly in December, along the coastal areas of Norway. Conversely, a stronger lagged correlation is observed between the winter NAO index and temperatures in northwest Sweden and coastal areas of Norway. Thus, NAO may influence both AMOC and winter temperatures (NAO drives both AMOC and temperatures). Furthermore, the paper investigates the impact of colder winters, whether caused by AMOC, NAO, or other factors like winds or sea ice changes, on electrical power and energy systems, highlighting potential challenges such as reduced electricity generation, increased electricity consumption, and the vulnerability of power grids to winter storms. The study concludes by emphasizing the importance of enhancing the knowledge of electrical engineering researchers regarding important climate indices, AMOC and NAO, the possible associations between them and winter temperatures, and addressing the challenges posed by the likelihood of colder winters in power systems.
Ključne besede: winter temperatures, biogas, Atlantic Meridional Overturning Circulation, AMOC, weaking, NAO, North Atlantic Oscillation, SSA, Singular Spectrum Analysis, electrical power and energy systems
Objavljeno v DKUM: 15.02.2024; Ogledov: 311; Prenosov: 46
Celotno besedilo (8,94 MB)
Gradivo ima več datotek! Več...