Generative Adversarial Network-based Data Recovery Method for Power Systems
Facing the problem of power system data loss, this paper proposes a power system data recovery method based on a generative adversarial network. The power system clustering method utilizes aggregated hierarchical clustering and takes into consideration the similarity between different power system d...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Sciendo
2024-01-01
|
| Series: | Applied Mathematics and Nonlinear Sciences |
| Subjects: | |
| Online Access: | https://doi.org/10.2478/amns-2024-0173 |
| _version_ | 1797303150786379776 |
|---|---|
| author | Yang Di Ji Ming Lv Yuntong Li Mengyu Gao Xuezhe |
| author_facet | Yang Di Ji Ming Lv Yuntong Li Mengyu Gao Xuezhe |
| author_sort | Yang Di |
| collection | DOAJ |
| description | Facing the problem of power system data loss, this paper proposes a power system data recovery method based on a generative adversarial network. The power system clustering method utilizes aggregated hierarchical clustering and takes into consideration the similarity between different power system data. To transform the power system data recovery problem into a data generation problem, an improved GAN network data analysis method is proposed that utilizes LSTM as a generator and discriminator. Through experimental tests, the LSTM-GAN method is tested with the LSTM method, interpolation method and low-rank method to compare its effect on lost data recovery under different signals of power system data static and dynamic and four fault scenarios. The results show that the root-mean-square errors of the LSTM-GAN method for recovering data under static-dynamic fluctuations are less than 1.2%, and the difference between the errors under 55% and 15% missing data conditions is only 0.77%, with the highest data recovery error of 2.32% in the power system fault scenarios. Therefore, the GAN-based power system data recovery method can effectively realize the recovery of lost data. |
| first_indexed | 2024-03-07T23:48:48Z |
| format | Article |
| id | doaj.art-1423876d4b9d49ae87ec7d715dc316b6 |
| institution | Directory Open Access Journal |
| issn | 2444-8656 |
| language | English |
| last_indexed | 2024-03-07T23:48:48Z |
| publishDate | 2024-01-01 |
| publisher | Sciendo |
| record_format | Article |
| series | Applied Mathematics and Nonlinear Sciences |
| spelling | doaj.art-1423876d4b9d49ae87ec7d715dc316b62024-02-19T09:03:35ZengSciendoApplied Mathematics and Nonlinear Sciences2444-86562024-01-019110.2478/amns-2024-0173Generative Adversarial Network-based Data Recovery Method for Power SystemsYang Di0Ji Ming1Lv Yuntong2Li Mengyu3Gao Xuezhe41State Grid Hebei Marketing Service Center, Shijiazhuang, Hebei, 050000, China.1State Grid Hebei Marketing Service Center, Shijiazhuang, Hebei, 050000, China.1State Grid Hebei Marketing Service Center, Shijiazhuang, Hebei, 050000, China.1State Grid Hebei Marketing Service Center, Shijiazhuang, Hebei, 050000, China.1State Grid Hebei Marketing Service Center, Shijiazhuang, Hebei, 050000, China.Facing the problem of power system data loss, this paper proposes a power system data recovery method based on a generative adversarial network. The power system clustering method utilizes aggregated hierarchical clustering and takes into consideration the similarity between different power system data. To transform the power system data recovery problem into a data generation problem, an improved GAN network data analysis method is proposed that utilizes LSTM as a generator and discriminator. Through experimental tests, the LSTM-GAN method is tested with the LSTM method, interpolation method and low-rank method to compare its effect on lost data recovery under different signals of power system data static and dynamic and four fault scenarios. The results show that the root-mean-square errors of the LSTM-GAN method for recovering data under static-dynamic fluctuations are less than 1.2%, and the difference between the errors under 55% and 15% missing data conditions is only 0.77%, with the highest data recovery error of 2.32% in the power system fault scenarios. Therefore, the GAN-based power system data recovery method can effectively realize the recovery of lost data.https://doi.org/10.2478/amns-2024-0173pmu measurement datalstm-ganhierarchical clusteringpower system clusteringsystem data recovery05c82 |
| spellingShingle | Yang Di Ji Ming Lv Yuntong Li Mengyu Gao Xuezhe Generative Adversarial Network-based Data Recovery Method for Power Systems Applied Mathematics and Nonlinear Sciences pmu measurement data lstm-gan hierarchical clustering power system clustering system data recovery 05c82 |
| title | Generative Adversarial Network-based Data Recovery Method for Power Systems |
| title_full | Generative Adversarial Network-based Data Recovery Method for Power Systems |
| title_fullStr | Generative Adversarial Network-based Data Recovery Method for Power Systems |
| title_full_unstemmed | Generative Adversarial Network-based Data Recovery Method for Power Systems |
| title_short | Generative Adversarial Network-based Data Recovery Method for Power Systems |
| title_sort | generative adversarial network based data recovery method for power systems |
| topic | pmu measurement data lstm-gan hierarchical clustering power system clustering system data recovery 05c82 |
| url | https://doi.org/10.2478/amns-2024-0173 |
| work_keys_str_mv | AT yangdi generativeadversarialnetworkbaseddatarecoverymethodforpowersystems AT jiming generativeadversarialnetworkbaseddatarecoverymethodforpowersystems AT lvyuntong generativeadversarialnetworkbaseddatarecoverymethodforpowersystems AT limengyu generativeadversarialnetworkbaseddatarecoverymethodforpowersystems AT gaoxuezhe generativeadversarialnetworkbaseddatarecoverymethodforpowersystems |