Fusion of MHSA and Boruta for key feature selection in power system transient angle stability
In response to challenges posed by existing transient stability feature selection methods, which often encounter limitations in searching for the optimum combination of critical features and lack an objective criterion for determining the optimal number of key features, this paper introduces a novel...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Department of Electric Power Engineering Technology
2025-01-01
|
| Series: | 电力工程技术 |
| Subjects: | |
| Online Access: | https://www.epet-info.com/dlgcjsen/article/abstract/231006225 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823865322991517696 |
|---|---|
| author | WANG Man ZHOU Xiaoyu CHEN Fan LAI Yening ZHU Ying |
| author_facet | WANG Man ZHOU Xiaoyu CHEN Fan LAI Yening ZHU Ying |
| author_sort | WANG Man |
| collection | DOAJ |
| description | In response to challenges posed by existing transient stability feature selection methods, which often encounter limitations in searching for the optimum combination of critical features and lack an objective criterion for determining the optimal number of key features, this paper introduces a novel approach. A transient power angle stability key feature selection method that seamlessly integrates multi-head self-attention (MHSA) and the Boruta algorithm. A deep neural network (DNN) with an MHSA model is initially constructed to execute transient stability assessments directly on the input grid features. The model dynamically adjusts attention weights during training, focusing on key features. Subsequently, the Boruta algorithm is employed to determine the number of key features. It generates a combination of real and virtual features, which the MHSA model trains to select the actual features that are higher than the maximum virtual feature weight, and the model autonomously determines the optimal number of key features. An analysis is conducted on the IEEE 39-node and 118-node systems to validate the proposed method. The results demonstrate that this approach ensures evaluation accuracy while significantly reducing the number of input features. Moreover, the key features identified exhibit higher evaluation accuracy than traditional methods. |
| format | Article |
| id | doaj-art-e43691468b80428d977b0a2543fae89c |
| institution | Kabale University |
| issn | 2096-3203 |
| language | zho |
| publishDate | 2025-01-01 |
| publisher | Editorial Department of Electric Power Engineering Technology |
| record_format | Article |
| series | 电力工程技术 |
| spelling | doaj-art-e43691468b80428d977b0a2543fae89c2025-02-08T08:40:18ZzhoEditorial Department of Electric Power Engineering Technology电力工程技术2096-32032025-01-0144115516410.12158/j.2096-3203.2025.01.016231006225Fusion of MHSA and Boruta for key feature selection in power system transient angle stabilityWANG Man0ZHOU Xiaoyu1CHEN Fan2LAI Yening3ZHU Ying4School of Electric Power Engineering, Nanjing Institute of Technology, Nanjing 211167, ChinaSchool of Electric Power Engineering, Nanjing Institute of Technology, Nanjing 211167, ChinaSchool of Electric Power Engineering, Nanjing Institute of Technology, Nanjing 211167, ChinaState Key Laboratory of Smart Grid Protection and Control, Nanjing 211106, ChinaCollege of Energy and Electrical Engineering, Hohai University, Nanjing 211100, ChinaIn response to challenges posed by existing transient stability feature selection methods, which often encounter limitations in searching for the optimum combination of critical features and lack an objective criterion for determining the optimal number of key features, this paper introduces a novel approach. A transient power angle stability key feature selection method that seamlessly integrates multi-head self-attention (MHSA) and the Boruta algorithm. A deep neural network (DNN) with an MHSA model is initially constructed to execute transient stability assessments directly on the input grid features. The model dynamically adjusts attention weights during training, focusing on key features. Subsequently, the Boruta algorithm is employed to determine the number of key features. It generates a combination of real and virtual features, which the MHSA model trains to select the actual features that are higher than the maximum virtual feature weight, and the model autonomously determines the optimal number of key features. An analysis is conducted on the IEEE 39-node and 118-node systems to validate the proposed method. The results demonstrate that this approach ensures evaluation accuracy while significantly reducing the number of input features. Moreover, the key features identified exhibit higher evaluation accuracy than traditional methods.https://www.epet-info.com/dlgcjsen/article/abstract/231006225multi-head self-attention (mhsa)boruta algorithmtransient stabilityfeature selectionkey featuresvirtual features |
| spellingShingle | WANG Man ZHOU Xiaoyu CHEN Fan LAI Yening ZHU Ying Fusion of MHSA and Boruta for key feature selection in power system transient angle stability 电力工程技术 multi-head self-attention (mhsa) boruta algorithm transient stability feature selection key features virtual features |
| title | Fusion of MHSA and Boruta for key feature selection in power system transient angle stability |
| title_full | Fusion of MHSA and Boruta for key feature selection in power system transient angle stability |
| title_fullStr | Fusion of MHSA and Boruta for key feature selection in power system transient angle stability |
| title_full_unstemmed | Fusion of MHSA and Boruta for key feature selection in power system transient angle stability |
| title_short | Fusion of MHSA and Boruta for key feature selection in power system transient angle stability |
| title_sort | fusion of mhsa and boruta for key feature selection in power system transient angle stability |
| topic | multi-head self-attention (mhsa) boruta algorithm transient stability feature selection key features virtual features |
| url | https://www.epet-info.com/dlgcjsen/article/abstract/231006225 |
| work_keys_str_mv | AT wangman fusionofmhsaandborutaforkeyfeatureselectioninpowersystemtransientanglestability AT zhouxiaoyu fusionofmhsaandborutaforkeyfeatureselectioninpowersystemtransientanglestability AT chenfan fusionofmhsaandborutaforkeyfeatureselectioninpowersystemtransientanglestability AT laiyening fusionofmhsaandborutaforkeyfeatureselectioninpowersystemtransientanglestability AT zhuying fusionofmhsaandborutaforkeyfeatureselectioninpowersystemtransientanglestability |