SlideShare una empresa de Scribd logo
1 de 12
Min-Woo Choi
Network Science Lab
Dept. of Artificial Intelligence
The Catholic University of Korea
E-mail: choimin1231@catholic.ac.kr
2023. 03. 06 Publish: 2023
Journal: Energy (IF: 11.45)
1
 Introduction
• Background
• Purpose
 Methodology
• Data description
• Experiment setup
 Results
 Conclusions
 Limitation and Future work
2
1. Introduction
Limitation
• Previous studies have focused on either physical models based on detailed atmospheric laws
or statistical approaches such as machine learning algorithms and time series analysis
techniques
• These methods are limited in their ability to capture the complex spatio-temporal correlations of
wind resources which can lead to inaccurate forecasts.
• The purpose of this paper is to present a generic framework for multi-step spatio-temporal
forecasting, using Graph Neural Networks (GNNs) and optional update functions such as
Transformer or LSTM networks.
Purpose of study
3
2. Methodology
Data description
• Location: Norwegian Meteorological Institute
• Period : June 23, 2015, and February 28, 2022
• Number of data: 14 out of 25 available stations
• Resolution: 10 min
• Variable: 8 feature (air temperature, air pressure, dew point,
relative humidity, wind direction, wind speed …)
• Pre-processing: If measurements for a single time-step
were missing for a particular station, linear interpolation
was used to fill the missing entries.
• Training/Validation/Test set: 60%/20%/20%
feature matrix, 𝑓𝑡 ∈ RNx8
4
2. Methodology
Experiment setup
• Look-back window of 32 time-steps was suitable for the 10-min and 1-h ahead forecasts
• Increased to 64 for the 4-h forecasts
Multilevel wavelet decomposition
5
2. Methodology
Fast Fourier Transformer
• FFTtransformer based on signal decomposition and an adapted attention mechanism,
named FFT-Attention.
• FFTransformer is comprised of two streams one which analyses signals with clear
periodicity and another that should learn trend components
• The basic configuration is the same as that of a original transformer, but to better
facilitate the analysis of periodic signals in the left-hand stream, we introduce the FFT-
Attention mechanism.
 FFT is applied to the key, query and value inputs
 Real and imaginary FFT outputs are concatenated with the frequency values, to provide
information on the corresponding frequency for the values in a particular position, similar to
the motivation behind positional encoding, before being fed to an MHA block.
• Outputs from the FFT-Attention module are then concatenated and projected, then
inverse FFT transform values back to the time domain.
6
2. Methodology
Spatial-temporal framework
 All models were constructed in an encoder-only setting, i.e. without
decoders, and used as update functions, 𝜙 (⋅) , in two-layer GNNs.
 Considering a node, 𝑖, its input features were 𝑣𝑖 ∈ R1×(𝑆+𝑃)×8
 𝑆 = historical look-back window
 𝑃 = Predict the next time-steps
7
2. Methodology
Benchmark models
• ST-MLP
• ST-LSTM
• ST-Transformer
• ST-LogSparse
• ST-Informer
• ST-Autoformer
• ST-FFTransformer
8
3. Results
Results of 10 min – 4 h prediction
9
3. Results
Graph connectivity
10
4. Conclusions
• The novel FFTransfomer was proposed which is based on signal decomposition using
wavelet transform and an adapted attention mechanism in the frequency domain - making
it significantly outperforming all other models for 4 hour ahead forecasts.
11
Thank you!

Más contenido relacionado

Similar a NS-CUK Joint Journal Club: Minwoo Choi, Review on "Spatio-Temporal Wind Speed Forecasting using Graph Networks and Novel Transformer Architectures", Appied Energy 333 2023

240219_RNN, LSTM code.pptxdddddddddddddddd
240219_RNN, LSTM code.pptxdddddddddddddddd240219_RNN, LSTM code.pptxdddddddddddddddd
240219_RNN, LSTM code.pptxddddddddddddddddssuser2624f71
 
ENERGY CONSUMPTION IMPROVEMENT OF TRADITIONAL CLUSTERING METHOD IN WIRELESS S...
ENERGY CONSUMPTION IMPROVEMENT OF TRADITIONAL CLUSTERING METHOD IN WIRELESS S...ENERGY CONSUMPTION IMPROVEMENT OF TRADITIONAL CLUSTERING METHOD IN WIRELESS S...
ENERGY CONSUMPTION IMPROVEMENT OF TRADITIONAL CLUSTERING METHOD IN WIRELESS S...IJCNCJournal
 
Direct digital frequency synthesizer
Direct digital frequency synthesizerDirect digital frequency synthesizer
Direct digital frequency synthesizerVenkat Malai Avichi
 
Comparison of Fine-tuning and Extension Strategies for Deep Convolutional Neu...
Comparison of Fine-tuning and Extension Strategies for Deep Convolutional Neu...Comparison of Fine-tuning and Extension Strategies for Deep Convolutional Neu...
Comparison of Fine-tuning and Extension Strategies for Deep Convolutional Neu...InVID Project
 
Cyclostationary analysis of polytime coded signals for lpi radars
Cyclostationary analysis of polytime coded signals for lpi radarsCyclostationary analysis of polytime coded signals for lpi radars
Cyclostationary analysis of polytime coded signals for lpi radarseSAT Journals
 
Efficiency of recurrent neural networks for seasonal trended time series mode...
Efficiency of recurrent neural networks for seasonal trended time series mode...Efficiency of recurrent neural networks for seasonal trended time series mode...
Efficiency of recurrent neural networks for seasonal trended time series mode...IJECEIAES
 
[20240429_LabSeminar_Huy]Spatio-Temporal Graph Neural Point Process for Traff...
[20240429_LabSeminar_Huy]Spatio-Temporal Graph Neural Point Process for Traff...[20240429_LabSeminar_Huy]Spatio-Temporal Graph Neural Point Process for Traff...
[20240429_LabSeminar_Huy]Spatio-Temporal Graph Neural Point Process for Traff...thanhdowork
 
A machine learning algorithm for classification of mental tasks.pdf
A machine learning algorithm for classification of mental tasks.pdfA machine learning algorithm for classification of mental tasks.pdf
A machine learning algorithm for classification of mental tasks.pdfPravinKshirsagar11
 
The ACTION Project: Applications Coordinate with Transport, IP and Optical Ne...
The ACTION Project: Applications Coordinate with Transport, IP and Optical Ne...The ACTION Project: Applications Coordinate with Transport, IP and Optical Ne...
The ACTION Project: Applications Coordinate with Transport, IP and Optical Ne...CPqD
 
Unified Framework for Learning Representation from EEG Data
Unified Framework for Learning Representation from EEG DataUnified Framework for Learning Representation from EEG Data
Unified Framework for Learning Representation from EEG DataFedEx Institute of Technology
 
A new method for controlling and maintaining
A new method for controlling and maintainingA new method for controlling and maintaining
A new method for controlling and maintainingIJCNCJournal
 
Voltage variations identification using Gabor Transform and rule-based classi...
Voltage variations identification using Gabor Transform and rule-based classi...Voltage variations identification using Gabor Transform and rule-based classi...
Voltage variations identification using Gabor Transform and rule-based classi...IJECEIAES
 
PERFORMANCE ANALYSIS OF RESOURCE SCHEDULING IN LTE FEMTOCELLS NETWORKS
PERFORMANCE ANALYSIS OF RESOURCE SCHEDULING IN LTE FEMTOCELLS NETWORKSPERFORMANCE ANALYSIS OF RESOURCE SCHEDULING IN LTE FEMTOCELLS NETWORKS
PERFORMANCE ANALYSIS OF RESOURCE SCHEDULING IN LTE FEMTOCELLS NETWORKScscpconf
 
Performance analysis of resource
Performance analysis of resourcePerformance analysis of resource
Performance analysis of resourcecsandit
 

Similar a NS-CUK Joint Journal Club: Minwoo Choi, Review on "Spatio-Temporal Wind Speed Forecasting using Graph Networks and Novel Transformer Architectures", Appied Energy 333 2023 (20)

240219_RNN, LSTM code.pptxdddddddddddddddd
240219_RNN, LSTM code.pptxdddddddddddddddd240219_RNN, LSTM code.pptxdddddddddddddddd
240219_RNN, LSTM code.pptxdddddddddddddddd
 
Class 27 signal processing techniques for the future smart grid.pdf
Class 27 signal processing techniques for the future smart grid.pdfClass 27 signal processing techniques for the future smart grid.pdf
Class 27 signal processing techniques for the future smart grid.pdf
 
STLF PPT jagdish singh
STLF PPT jagdish singhSTLF PPT jagdish singh
STLF PPT jagdish singh
 
ENERGY CONSUMPTION IMPROVEMENT OF TRADITIONAL CLUSTERING METHOD IN WIRELESS S...
ENERGY CONSUMPTION IMPROVEMENT OF TRADITIONAL CLUSTERING METHOD IN WIRELESS S...ENERGY CONSUMPTION IMPROVEMENT OF TRADITIONAL CLUSTERING METHOD IN WIRELESS S...
ENERGY CONSUMPTION IMPROVEMENT OF TRADITIONAL CLUSTERING METHOD IN WIRELESS S...
 
Direct digital frequency synthesizer
Direct digital frequency synthesizerDirect digital frequency synthesizer
Direct digital frequency synthesizer
 
2
22
2
 
Comparison of Fine-tuning and Extension Strategies for Deep Convolutional Neu...
Comparison of Fine-tuning and Extension Strategies for Deep Convolutional Neu...Comparison of Fine-tuning and Extension Strategies for Deep Convolutional Neu...
Comparison of Fine-tuning and Extension Strategies for Deep Convolutional Neu...
 
Cyclostationary analysis of polytime coded signals for lpi radars
Cyclostationary analysis of polytime coded signals for lpi radarsCyclostationary analysis of polytime coded signals for lpi radars
Cyclostationary analysis of polytime coded signals for lpi radars
 
Efficiency of recurrent neural networks for seasonal trended time series mode...
Efficiency of recurrent neural networks for seasonal trended time series mode...Efficiency of recurrent neural networks for seasonal trended time series mode...
Efficiency of recurrent neural networks for seasonal trended time series mode...
 
MU- mimo [autosaved]
MU- mimo [autosaved]MU- mimo [autosaved]
MU- mimo [autosaved]
 
D0542130
D0542130D0542130
D0542130
 
[20240429_LabSeminar_Huy]Spatio-Temporal Graph Neural Point Process for Traff...
[20240429_LabSeminar_Huy]Spatio-Temporal Graph Neural Point Process for Traff...[20240429_LabSeminar_Huy]Spatio-Temporal Graph Neural Point Process for Traff...
[20240429_LabSeminar_Huy]Spatio-Temporal Graph Neural Point Process for Traff...
 
A machine learning algorithm for classification of mental tasks.pdf
A machine learning algorithm for classification of mental tasks.pdfA machine learning algorithm for classification of mental tasks.pdf
A machine learning algorithm for classification of mental tasks.pdf
 
The ACTION Project: Applications Coordinate with Transport, IP and Optical Ne...
The ACTION Project: Applications Coordinate with Transport, IP and Optical Ne...The ACTION Project: Applications Coordinate with Transport, IP and Optical Ne...
The ACTION Project: Applications Coordinate with Transport, IP and Optical Ne...
 
Yeasin
YeasinYeasin
Yeasin
 
Unified Framework for Learning Representation from EEG Data
Unified Framework for Learning Representation from EEG DataUnified Framework for Learning Representation from EEG Data
Unified Framework for Learning Representation from EEG Data
 
A new method for controlling and maintaining
A new method for controlling and maintainingA new method for controlling and maintaining
A new method for controlling and maintaining
 
Voltage variations identification using Gabor Transform and rule-based classi...
Voltage variations identification using Gabor Transform and rule-based classi...Voltage variations identification using Gabor Transform and rule-based classi...
Voltage variations identification using Gabor Transform and rule-based classi...
 
PERFORMANCE ANALYSIS OF RESOURCE SCHEDULING IN LTE FEMTOCELLS NETWORKS
PERFORMANCE ANALYSIS OF RESOURCE SCHEDULING IN LTE FEMTOCELLS NETWORKSPERFORMANCE ANALYSIS OF RESOURCE SCHEDULING IN LTE FEMTOCELLS NETWORKS
PERFORMANCE ANALYSIS OF RESOURCE SCHEDULING IN LTE FEMTOCELLS NETWORKS
 
Performance analysis of resource
Performance analysis of resourcePerformance analysis of resource
Performance analysis of resource
 

Más de ssuser4b1f48

NS-CUK Seminar: V.T.Hoang, Review on "GOAT: A Global Transformer on Large-sca...
NS-CUK Seminar: V.T.Hoang, Review on "GOAT: A Global Transformer on Large-sca...NS-CUK Seminar: V.T.Hoang, Review on "GOAT: A Global Transformer on Large-sca...
NS-CUK Seminar: V.T.Hoang, Review on "GOAT: A Global Transformer on Large-sca...ssuser4b1f48
 
NS-CUK Seminar: J.H.Lee, Review on "Graph Propagation Transformer for Graph R...
NS-CUK Seminar: J.H.Lee, Review on "Graph Propagation Transformer for Graph R...NS-CUK Seminar: J.H.Lee, Review on "Graph Propagation Transformer for Graph R...
NS-CUK Seminar: J.H.Lee, Review on "Graph Propagation Transformer for Graph R...ssuser4b1f48
 
NS-CUK Seminar: H.B.Kim, Review on "Cluster-GCN: An Efficient Algorithm for ...
NS-CUK Seminar: H.B.Kim,  Review on "Cluster-GCN: An Efficient Algorithm for ...NS-CUK Seminar: H.B.Kim,  Review on "Cluster-GCN: An Efficient Algorithm for ...
NS-CUK Seminar: H.B.Kim, Review on "Cluster-GCN: An Efficient Algorithm for ...ssuser4b1f48
 
NS-CUK Seminar: H.E.Lee, Review on "Weisfeiler and Leman Go Neural: Higher-O...
NS-CUK Seminar: H.E.Lee,  Review on "Weisfeiler and Leman Go Neural: Higher-O...NS-CUK Seminar: H.E.Lee,  Review on "Weisfeiler and Leman Go Neural: Higher-O...
NS-CUK Seminar: H.E.Lee, Review on "Weisfeiler and Leman Go Neural: Higher-O...ssuser4b1f48
 
NS-CUK Seminar:V.T.Hoang, Review on "GRPE: Relative Positional Encoding for G...
NS-CUK Seminar:V.T.Hoang, Review on "GRPE: Relative Positional Encoding for G...NS-CUK Seminar:V.T.Hoang, Review on "GRPE: Relative Positional Encoding for G...
NS-CUK Seminar:V.T.Hoang, Review on "GRPE: Relative Positional Encoding for G...ssuser4b1f48
 
NS-CUK Seminar: J.H.Lee, Review on "Learnable Structural Semantic Readout for...
NS-CUK Seminar: J.H.Lee, Review on "Learnable Structural Semantic Readout for...NS-CUK Seminar: J.H.Lee, Review on "Learnable Structural Semantic Readout for...
NS-CUK Seminar: J.H.Lee, Review on "Learnable Structural Semantic Readout for...ssuser4b1f48
 
Aug 22nd, 2023: Case Studies - The Art and Science of Animation Production)
Aug 22nd, 2023: Case Studies - The Art and Science of Animation Production)Aug 22nd, 2023: Case Studies - The Art and Science of Animation Production)
Aug 22nd, 2023: Case Studies - The Art and Science of Animation Production)ssuser4b1f48
 
Aug 17th, 2023: Case Studies - Examining Gamification through Virtual/Augment...
Aug 17th, 2023: Case Studies - Examining Gamification through Virtual/Augment...Aug 17th, 2023: Case Studies - Examining Gamification through Virtual/Augment...
Aug 17th, 2023: Case Studies - Examining Gamification through Virtual/Augment...ssuser4b1f48
 
Aug 10th, 2023: Case Studies - The Power of eXtended Reality (XR) with 360°
Aug 10th, 2023: Case Studies - The Power of eXtended Reality (XR) with 360°Aug 10th, 2023: Case Studies - The Power of eXtended Reality (XR) with 360°
Aug 10th, 2023: Case Studies - The Power of eXtended Reality (XR) with 360°ssuser4b1f48
 
Aug 8th, 2023: Case Studies - Utilizing eXtended Reality (XR) in Drones)
Aug 8th, 2023: Case Studies - Utilizing eXtended Reality (XR) in Drones)Aug 8th, 2023: Case Studies - Utilizing eXtended Reality (XR) in Drones)
Aug 8th, 2023: Case Studies - Utilizing eXtended Reality (XR) in Drones)ssuser4b1f48
 
NS-CUK Seminar: J.H.Lee, Review on "Learnable Structural Semantic Readout for...
NS-CUK Seminar: J.H.Lee, Review on "Learnable Structural Semantic Readout for...NS-CUK Seminar: J.H.Lee, Review on "Learnable Structural Semantic Readout for...
NS-CUK Seminar: J.H.Lee, Review on "Learnable Structural Semantic Readout for...ssuser4b1f48
 
NS-CUK Seminar: H.E.Lee, Review on "Gated Graph Sequence Neural Networks", I...
NS-CUK Seminar: H.E.Lee,  Review on "Gated Graph Sequence Neural Networks", I...NS-CUK Seminar: H.E.Lee,  Review on "Gated Graph Sequence Neural Networks", I...
NS-CUK Seminar: H.E.Lee, Review on "Gated Graph Sequence Neural Networks", I...ssuser4b1f48
 
NS-CUK Seminar:V.T.Hoang, Review on "Augmentation-Free Self-Supervised Learni...
NS-CUK Seminar:V.T.Hoang, Review on "Augmentation-Free Self-Supervised Learni...NS-CUK Seminar:V.T.Hoang, Review on "Augmentation-Free Self-Supervised Learni...
NS-CUK Seminar:V.T.Hoang, Review on "Augmentation-Free Self-Supervised Learni...ssuser4b1f48
 
NS-CUK Journal club: H.E.Lee, Review on " A biomedical knowledge graph-based ...
NS-CUK Journal club: H.E.Lee, Review on " A biomedical knowledge graph-based ...NS-CUK Journal club: H.E.Lee, Review on " A biomedical knowledge graph-based ...
NS-CUK Journal club: H.E.Lee, Review on " A biomedical knowledge graph-based ...ssuser4b1f48
 
NS-CUK Seminar: H.E.Lee, Review on "PTE: Predictive Text Embedding through L...
NS-CUK Seminar: H.E.Lee,  Review on "PTE: Predictive Text Embedding through L...NS-CUK Seminar: H.E.Lee,  Review on "PTE: Predictive Text Embedding through L...
NS-CUK Seminar: H.E.Lee, Review on "PTE: Predictive Text Embedding through L...ssuser4b1f48
 
NS-CUK Seminar: H.B.Kim, Review on "Inductive Representation Learning on Lar...
NS-CUK Seminar: H.B.Kim,  Review on "Inductive Representation Learning on Lar...NS-CUK Seminar: H.B.Kim,  Review on "Inductive Representation Learning on Lar...
NS-CUK Seminar: H.B.Kim, Review on "Inductive Representation Learning on Lar...ssuser4b1f48
 
NS-CUK Seminar: H.E.Lee, Review on "PTE: Predictive Text Embedding through L...
NS-CUK Seminar: H.E.Lee,  Review on "PTE: Predictive Text Embedding through L...NS-CUK Seminar: H.E.Lee,  Review on "PTE: Predictive Text Embedding through L...
NS-CUK Seminar: H.E.Lee, Review on "PTE: Predictive Text Embedding through L...ssuser4b1f48
 
NS-CUK Seminar: J.H.Lee, Review on "Relational Self-Supervised Learning on Gr...
NS-CUK Seminar: J.H.Lee, Review on "Relational Self-Supervised Learning on Gr...NS-CUK Seminar: J.H.Lee, Review on "Relational Self-Supervised Learning on Gr...
NS-CUK Seminar: J.H.Lee, Review on "Relational Self-Supervised Learning on Gr...ssuser4b1f48
 
NS-CUK Seminar: H.B.Kim, Review on "metapath2vec: Scalable representation le...
NS-CUK Seminar: H.B.Kim,  Review on "metapath2vec: Scalable representation le...NS-CUK Seminar: H.B.Kim,  Review on "metapath2vec: Scalable representation le...
NS-CUK Seminar: H.B.Kim, Review on "metapath2vec: Scalable representation le...ssuser4b1f48
 
NS-CUK Seminar: H.E.Lee, Review on "Graph Star Net for Generalized Multi-Tas...
NS-CUK Seminar: H.E.Lee,  Review on "Graph Star Net for Generalized Multi-Tas...NS-CUK Seminar: H.E.Lee,  Review on "Graph Star Net for Generalized Multi-Tas...
NS-CUK Seminar: H.E.Lee, Review on "Graph Star Net for Generalized Multi-Tas...ssuser4b1f48
 

Más de ssuser4b1f48 (20)

NS-CUK Seminar: V.T.Hoang, Review on "GOAT: A Global Transformer on Large-sca...
NS-CUK Seminar: V.T.Hoang, Review on "GOAT: A Global Transformer on Large-sca...NS-CUK Seminar: V.T.Hoang, Review on "GOAT: A Global Transformer on Large-sca...
NS-CUK Seminar: V.T.Hoang, Review on "GOAT: A Global Transformer on Large-sca...
 
NS-CUK Seminar: J.H.Lee, Review on "Graph Propagation Transformer for Graph R...
NS-CUK Seminar: J.H.Lee, Review on "Graph Propagation Transformer for Graph R...NS-CUK Seminar: J.H.Lee, Review on "Graph Propagation Transformer for Graph R...
NS-CUK Seminar: J.H.Lee, Review on "Graph Propagation Transformer for Graph R...
 
NS-CUK Seminar: H.B.Kim, Review on "Cluster-GCN: An Efficient Algorithm for ...
NS-CUK Seminar: H.B.Kim,  Review on "Cluster-GCN: An Efficient Algorithm for ...NS-CUK Seminar: H.B.Kim,  Review on "Cluster-GCN: An Efficient Algorithm for ...
NS-CUK Seminar: H.B.Kim, Review on "Cluster-GCN: An Efficient Algorithm for ...
 
NS-CUK Seminar: H.E.Lee, Review on "Weisfeiler and Leman Go Neural: Higher-O...
NS-CUK Seminar: H.E.Lee,  Review on "Weisfeiler and Leman Go Neural: Higher-O...NS-CUK Seminar: H.E.Lee,  Review on "Weisfeiler and Leman Go Neural: Higher-O...
NS-CUK Seminar: H.E.Lee, Review on "Weisfeiler and Leman Go Neural: Higher-O...
 
NS-CUK Seminar:V.T.Hoang, Review on "GRPE: Relative Positional Encoding for G...
NS-CUK Seminar:V.T.Hoang, Review on "GRPE: Relative Positional Encoding for G...NS-CUK Seminar:V.T.Hoang, Review on "GRPE: Relative Positional Encoding for G...
NS-CUK Seminar:V.T.Hoang, Review on "GRPE: Relative Positional Encoding for G...
 
NS-CUK Seminar: J.H.Lee, Review on "Learnable Structural Semantic Readout for...
NS-CUK Seminar: J.H.Lee, Review on "Learnable Structural Semantic Readout for...NS-CUK Seminar: J.H.Lee, Review on "Learnable Structural Semantic Readout for...
NS-CUK Seminar: J.H.Lee, Review on "Learnable Structural Semantic Readout for...
 
Aug 22nd, 2023: Case Studies - The Art and Science of Animation Production)
Aug 22nd, 2023: Case Studies - The Art and Science of Animation Production)Aug 22nd, 2023: Case Studies - The Art and Science of Animation Production)
Aug 22nd, 2023: Case Studies - The Art and Science of Animation Production)
 
Aug 17th, 2023: Case Studies - Examining Gamification through Virtual/Augment...
Aug 17th, 2023: Case Studies - Examining Gamification through Virtual/Augment...Aug 17th, 2023: Case Studies - Examining Gamification through Virtual/Augment...
Aug 17th, 2023: Case Studies - Examining Gamification through Virtual/Augment...
 
Aug 10th, 2023: Case Studies - The Power of eXtended Reality (XR) with 360°
Aug 10th, 2023: Case Studies - The Power of eXtended Reality (XR) with 360°Aug 10th, 2023: Case Studies - The Power of eXtended Reality (XR) with 360°
Aug 10th, 2023: Case Studies - The Power of eXtended Reality (XR) with 360°
 
Aug 8th, 2023: Case Studies - Utilizing eXtended Reality (XR) in Drones)
Aug 8th, 2023: Case Studies - Utilizing eXtended Reality (XR) in Drones)Aug 8th, 2023: Case Studies - Utilizing eXtended Reality (XR) in Drones)
Aug 8th, 2023: Case Studies - Utilizing eXtended Reality (XR) in Drones)
 
NS-CUK Seminar: J.H.Lee, Review on "Learnable Structural Semantic Readout for...
NS-CUK Seminar: J.H.Lee, Review on "Learnable Structural Semantic Readout for...NS-CUK Seminar: J.H.Lee, Review on "Learnable Structural Semantic Readout for...
NS-CUK Seminar: J.H.Lee, Review on "Learnable Structural Semantic Readout for...
 
NS-CUK Seminar: H.E.Lee, Review on "Gated Graph Sequence Neural Networks", I...
NS-CUK Seminar: H.E.Lee,  Review on "Gated Graph Sequence Neural Networks", I...NS-CUK Seminar: H.E.Lee,  Review on "Gated Graph Sequence Neural Networks", I...
NS-CUK Seminar: H.E.Lee, Review on "Gated Graph Sequence Neural Networks", I...
 
NS-CUK Seminar:V.T.Hoang, Review on "Augmentation-Free Self-Supervised Learni...
NS-CUK Seminar:V.T.Hoang, Review on "Augmentation-Free Self-Supervised Learni...NS-CUK Seminar:V.T.Hoang, Review on "Augmentation-Free Self-Supervised Learni...
NS-CUK Seminar:V.T.Hoang, Review on "Augmentation-Free Self-Supervised Learni...
 
NS-CUK Journal club: H.E.Lee, Review on " A biomedical knowledge graph-based ...
NS-CUK Journal club: H.E.Lee, Review on " A biomedical knowledge graph-based ...NS-CUK Journal club: H.E.Lee, Review on " A biomedical knowledge graph-based ...
NS-CUK Journal club: H.E.Lee, Review on " A biomedical knowledge graph-based ...
 
NS-CUK Seminar: H.E.Lee, Review on "PTE: Predictive Text Embedding through L...
NS-CUK Seminar: H.E.Lee,  Review on "PTE: Predictive Text Embedding through L...NS-CUK Seminar: H.E.Lee,  Review on "PTE: Predictive Text Embedding through L...
NS-CUK Seminar: H.E.Lee, Review on "PTE: Predictive Text Embedding through L...
 
NS-CUK Seminar: H.B.Kim, Review on "Inductive Representation Learning on Lar...
NS-CUK Seminar: H.B.Kim,  Review on "Inductive Representation Learning on Lar...NS-CUK Seminar: H.B.Kim,  Review on "Inductive Representation Learning on Lar...
NS-CUK Seminar: H.B.Kim, Review on "Inductive Representation Learning on Lar...
 
NS-CUK Seminar: H.E.Lee, Review on "PTE: Predictive Text Embedding through L...
NS-CUK Seminar: H.E.Lee,  Review on "PTE: Predictive Text Embedding through L...NS-CUK Seminar: H.E.Lee,  Review on "PTE: Predictive Text Embedding through L...
NS-CUK Seminar: H.E.Lee, Review on "PTE: Predictive Text Embedding through L...
 
NS-CUK Seminar: J.H.Lee, Review on "Relational Self-Supervised Learning on Gr...
NS-CUK Seminar: J.H.Lee, Review on "Relational Self-Supervised Learning on Gr...NS-CUK Seminar: J.H.Lee, Review on "Relational Self-Supervised Learning on Gr...
NS-CUK Seminar: J.H.Lee, Review on "Relational Self-Supervised Learning on Gr...
 
NS-CUK Seminar: H.B.Kim, Review on "metapath2vec: Scalable representation le...
NS-CUK Seminar: H.B.Kim,  Review on "metapath2vec: Scalable representation le...NS-CUK Seminar: H.B.Kim,  Review on "metapath2vec: Scalable representation le...
NS-CUK Seminar: H.B.Kim, Review on "metapath2vec: Scalable representation le...
 
NS-CUK Seminar: H.E.Lee, Review on "Graph Star Net for Generalized Multi-Tas...
NS-CUK Seminar: H.E.Lee,  Review on "Graph Star Net for Generalized Multi-Tas...NS-CUK Seminar: H.E.Lee,  Review on "Graph Star Net for Generalized Multi-Tas...
NS-CUK Seminar: H.E.Lee, Review on "Graph Star Net for Generalized Multi-Tas...
 

Último

04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptxHampshireHUG
 
Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024The Digital Insurer
 
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUK Journal
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slidespraypatel2
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityPrincipled Technologies
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slidevu2urc
 
Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfsudhanshuwaghmare1
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CVKhem
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024Rafal Los
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...Martijn de Jong
 
What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?Antenna Manufacturer Coco
 
Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...Enterprise Knowledge
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdfhans926745
 
Artificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and MythsArtificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and MythsJoaquim Jorge
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking MenDelhi Call girls
 
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...Neo4j
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsMaria Levchenko
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)Gabriella Davis
 
CNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of ServiceCNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of Servicegiselly40
 

Último (20)

04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
 
Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024Tata AIG General Insurance Company - Insurer Innovation Award 2024
Tata AIG General Insurance Company - Insurer Innovation Award 2024
 
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdfUnderstanding Discord NSFW Servers A Guide for Responsible Users.pdf
Understanding Discord NSFW Servers A Guide for Responsible Users.pdf
 
Slack Application Development 101 Slides
Slack Application Development 101 SlidesSlack Application Development 101 Slides
Slack Application Development 101 Slides
 
Boost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivityBoost PC performance: How more available memory can improve productivity
Boost PC performance: How more available memory can improve productivity
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
 
Boost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdfBoost Fertility New Invention Ups Success Rates.pdf
Boost Fertility New Invention Ups Success Rates.pdf
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CV
 
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law DevelopmentsTrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
TrustArc Webinar - Stay Ahead of US State Data Privacy Law Developments
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...
 
What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?What Are The Drone Anti-jamming Systems Technology?
What Are The Drone Anti-jamming Systems Technology?
 
Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...Driving Behavioral Change for Information Management through Data-Driven Gree...
Driving Behavioral Change for Information Management through Data-Driven Gree...
 
[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf[2024]Digital Global Overview Report 2024 Meltwater.pdf
[2024]Digital Global Overview Report 2024 Meltwater.pdf
 
Artificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and MythsArtificial Intelligence: Facts and Myths
Artificial Intelligence: Facts and Myths
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed texts
 
A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)A Domino Admins Adventures (Engage 2024)
A Domino Admins Adventures (Engage 2024)
 
CNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of ServiceCNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of Service
 

NS-CUK Joint Journal Club: Minwoo Choi, Review on "Spatio-Temporal Wind Speed Forecasting using Graph Networks and Novel Transformer Architectures", Appied Energy 333 2023

  • 1. Min-Woo Choi Network Science Lab Dept. of Artificial Intelligence The Catholic University of Korea E-mail: choimin1231@catholic.ac.kr 2023. 03. 06 Publish: 2023 Journal: Energy (IF: 11.45)
  • 2. 1  Introduction • Background • Purpose  Methodology • Data description • Experiment setup  Results  Conclusions  Limitation and Future work
  • 3. 2 1. Introduction Limitation • Previous studies have focused on either physical models based on detailed atmospheric laws or statistical approaches such as machine learning algorithms and time series analysis techniques • These methods are limited in their ability to capture the complex spatio-temporal correlations of wind resources which can lead to inaccurate forecasts. • The purpose of this paper is to present a generic framework for multi-step spatio-temporal forecasting, using Graph Neural Networks (GNNs) and optional update functions such as Transformer or LSTM networks. Purpose of study
  • 4. 3 2. Methodology Data description • Location: Norwegian Meteorological Institute • Period : June 23, 2015, and February 28, 2022 • Number of data: 14 out of 25 available stations • Resolution: 10 min • Variable: 8 feature (air temperature, air pressure, dew point, relative humidity, wind direction, wind speed …) • Pre-processing: If measurements for a single time-step were missing for a particular station, linear interpolation was used to fill the missing entries. • Training/Validation/Test set: 60%/20%/20% feature matrix, 𝑓𝑡 ∈ RNx8
  • 5. 4 2. Methodology Experiment setup • Look-back window of 32 time-steps was suitable for the 10-min and 1-h ahead forecasts • Increased to 64 for the 4-h forecasts Multilevel wavelet decomposition
  • 6. 5 2. Methodology Fast Fourier Transformer • FFTtransformer based on signal decomposition and an adapted attention mechanism, named FFT-Attention. • FFTransformer is comprised of two streams one which analyses signals with clear periodicity and another that should learn trend components • The basic configuration is the same as that of a original transformer, but to better facilitate the analysis of periodic signals in the left-hand stream, we introduce the FFT- Attention mechanism.  FFT is applied to the key, query and value inputs  Real and imaginary FFT outputs are concatenated with the frequency values, to provide information on the corresponding frequency for the values in a particular position, similar to the motivation behind positional encoding, before being fed to an MHA block. • Outputs from the FFT-Attention module are then concatenated and projected, then inverse FFT transform values back to the time domain.
  • 7. 6 2. Methodology Spatial-temporal framework  All models were constructed in an encoder-only setting, i.e. without decoders, and used as update functions, 𝜙 (⋅) , in two-layer GNNs.  Considering a node, 𝑖, its input features were 𝑣𝑖 ∈ R1×(𝑆+𝑃)×8  𝑆 = historical look-back window  𝑃 = Predict the next time-steps
  • 8. 7 2. Methodology Benchmark models • ST-MLP • ST-LSTM • ST-Transformer • ST-LogSparse • ST-Informer • ST-Autoformer • ST-FFTransformer
  • 9. 8 3. Results Results of 10 min – 4 h prediction
  • 11. 10 4. Conclusions • The novel FFTransfomer was proposed which is based on signal decomposition using wavelet transform and an adapted attention mechanism in the frequency domain - making it significantly outperforming all other models for 4 hour ahead forecasts.

Notas del editor

  1. 안녕하십니까, 저는 기상환경 관련 연구를 수행했던 최민우라고 하구요. 저의 전공지식을 토대로 이오준 교수님과 앞으로 연구를 진행해 나갈 예정입니다. 앞으로 영어 실력의 향상을 위해 영어로 발표하겠으나. 아직은 대본을 대부분 참조한다는 점 양해부탁드립니다. The topic of my presentation is Short-term wind speed forecasting based on spatial-temporal graph transformer networks.
  2. The order of contents is as follows:
  3. First, the limitations of previous studies in wind speed prediction are as follows. When we use a spatiotemporal prediction model to predict wind speed based on CNN, a square array or nodes with regular intervals are required. Like that grid. However, in actually it is not composed of regular data. Therefore, as a related study, Fu extracted spatial correlation using the STAN model. and Kyodayar extracted spatiotemporal information of wind speed using the GCDLA model. There are few related studies, but taking advantage of related studies, this study proposed STGTN to predict wind speed.
  4. The data used were wind farms located in Danish offshore, and data were collected at 10-minute intervals with 111 wind turbine nodes. The data set was divided into 3:1:1, and the wind speed was predicted using the historical data of each 12 points.
  5. Benchmark models were constructed to validate the proposed model. First, SVR model not include the spatial information, DL-STF and STAN models based on spatial-temporal information and the STGTN-T model using Transformer was used instead of MLP model.
  6. The ST-LogSparse and ST-Informer performed consistently better than the ST-Transformer model across all forecasting horizons in terms of both MSE and MAE, which showed the potential improvements brought by the ProbSparse and convolutional attention mechanisms for wind forecasting. In general, all Transformer-based models attained better results ST-Autoformer model performed very well for the 1- and 6-step forecasts, its performance was seen to degrade for the 24-step setting, where it was inferior to all other Transformerbased models. Both the ST-FFTransformer and ST-Autoformer achieved the best accuracy for three settings each, likely because of data decomposition. One reason for the superior performance of the FFTransformer architecture for the longer forecasting setting might be that it separately processes the frequency and trend components, using the two streams described in Section 4.2, with attention and feedforward modules applied to both. On the other hand, the Autoformer model is mainly focused on processing periodic components using the Auto-Correlation module and does not perform significant processing on extracted trend components after series decomposition, making it potentially struggle to learn longer-term trends that lack periodicity. Even though the ST-FFTransformer achieved good results across different horizons, it did not outperform the ST-Autoformer in the short-term. This was despite more advanced decomposition using MDWD, which was initially thought better at extracting trend and periodic components at different frequencies, compared to the simple moving average operation used in the Autoformer. In terms of decomposition, this could indicate that the MDWD is not necessarily superior to the ‘Series Decomposition’, or more likely, that the repeated use of decomposition in the Autoformer might work slightly better than the single decomposition performed on the inputs in the ST-FFTransformer.
  7. Even though this study did not focus on discussing better connectivity strategies for wind forecasting or using learnable adjacency matrices, we conduct a brief investigation into whether some of the connections could potentially be removed. The sharp decrease indicated that the models were able to successfully leverage spatial correlations to improve forecasts, proving the effectiveness of the proposed GNN architectures. Nevertheless, since MAEs converged to constant values for non-complete graphs, it indicated that a number of connections could potentially be removed without impairing predictive performance. Further work would therefore be desirable to investigate better methods for which to construct the graphs. or learn optimal connectivity for spatio-temporal wind forecasting. spatial correlations were thought less important than for the 1- and 4-h forecasts, due to the large physical distances between nodes resulting in wind fields not having time to propagate in the immediate short-term.
  8. The proposed model is stable and has excellent spatio-temporal predictability.
  9. Anyone have any questions?