Applying Deep Learning-based concepts for the detection of device misconfigurations in power systems

Fellner, David; Strasser, Thomas; Kästner, Wolfgang

doi:10.1016/j.segan.2022.100851

Cited by 10 publications

(4 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…2) Combining Recurrent and Attention: R-Transformer [386] inherits the Transformers' architecture and is adding what they call "Local RNN" to capture sequential information in data. The main improvement proposed is defining a sequence window to capture the sequential information and sliding the Local RNN over the whole time series to get the global sequential information [387]. This approach is similar to 1-D CNN; however, CNN ignores the sequential information of positions.…”

Section: ) Rnn Vs Transformermentioning

confidence: 99%

Deep Representation Learning: Fundamentals, Technologies, Applications, and Open Challenges

Payandeh,

Baghaei,

Fayyazsanavi

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Machine learning algorithms have had a profound impact on the field of computer science over the past few decades. The performance of these algorithms heavily depends on the representations derived from the data during the learning process. Successful learning processes aim to produce concise, discrete, meaningful representations that can be effectively applied to various tasks. Recent advancements in deep learning models have proven to be highly effective in capturing high-dimensional, non-linear, and multi-modal characteristics. In this work, we provide a comprehensive overview of the current state-of-the-art in deep representation learning and the principles and developments made in the process of representation learning. Our study encompasses both supervised and unsupervised methods, including popular techniques such as autoencoders, self-supervised methods, and deep neural networks. Furthermore, we explore a wide range of applications, including image recognition and natural language processing. In addition, we discuss recent trends, key issues, and open challenges in the field. This survey endeavors to make a significant contribution to the field of deep representation learning, fostering its understanding and facilitating further advancements.

show abstract

Section: ) Rnn Vs Transformermentioning

confidence: 99%

Deep Representation Learning: Fundamentals, Technologies, Applications, and Open Challenges

Payandeh,

Baghaei,

Fayyazsanavi

et al. 2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The assessment of detection approaches on the device level revealed the so-called R Transformer approach to be the best-performing method [8]. This approach, which is sketched in Figure 4, uses Recurrent Neural Networks (RNN) to capture time dependencies in the data on a local scale.…”

Section: Figure 3: Inverter Configurations To Be Detectedmentioning

confidence: 99%

Data Driven Transformer Level Misconfiguration Detection in Power Distribution Grids

Fellner

Strasser

Kästner

et al. 2022

2022 IEEE International Conference on Systems, Man, and Cybernetics (SMC)

View full text Add to dashboard Cite

As more novel devices are integrated into the electricity grid due to the changes taking place in the energy system, ways of detecting deviations from the intended settings are needed. If misconfigurations of, for example, reactive power control curves of inverters go unnoticed, the safe and reliable operation of the power grid can no longer be ensured due to possible voltage violations or overloadings. Therefore, methods of detection of misconfigurations of said inverters using operational data at transformers are presented and compared. These methods include preprocessing by dimensionality reduction as well as detection by supervised learning approaches. The data used is of high reliability as it was collected in a lab setting reenacting typical and relevant grid operation situations. Furthermore, this data was recreated by simulation to validate the simulation data, which could also potentially be used for detection applications on a bigger scale. The results for both data sources were compared and conclusions drawn about applicability and usability for grid operators.

show abstract

“…Literature Review: The popularity of deep learning in the field of power systems has been well-documented in recent research [8,9,10], with several studies showing that it can be used to learn the solutions to computationally intensive power system analysis algorithms. In [11], the authors used a combination of recurrent and feed-forward neural networks to solve the power system SE problem using measurement data and the history of network voltages.…”

Section: Introductionmentioning

confidence: 99%

Graph neural networks on factor graphs for robust, fast, and scalable linear state estimation with PMUs

Ognjen¹,

Cosovic

Miskovic³

et al. 2023

Sustainable Energy, Grids and Networks

View full text Add to dashboard Cite

Applying Deep Learning-based concepts for the detection of device misconfigurations in power systems

Cited by 10 publications

References 18 publications

Deep Representation Learning: Fundamentals, Technologies, Applications, and Open Challenges

Deep Representation Learning: Fundamentals, Technologies, Applications, and Open Challenges

Data Driven Transformer Level Misconfiguration Detection in Power Distribution Grids

Graph neural networks on factor graphs for robust, fast, and scalable linear state estimation with PMUs

Contact Info

Product

Resources

About