Anomaly detection of smart grid stealing network attacks based on deep autoencoder
Existing anomaly detectors in AMIs suffer from shallow architectures, which impede their ability to capture temporal correlations and complex patterns in electricity consumption data, thus impact detection performance adversely. A deep (stacked) autoencoder structure based on Long Short-Term Memory...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
National Computer System Engineering Research Institute of China
2024-02-01
|
| Series: | Dianzi Jishu Yingyong |
| Subjects: | |
| Online Access: | http://www.chinaaet.com/article/3000163482 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849423467310481408 |
|---|---|
| author | Huang Yan Li Jincan Yang Xiaqin Li Pei Li Zi |
| author_facet | Huang Yan Li Jincan Yang Xiaqin Li Pei Li Zi |
| author_sort | Huang Yan |
| collection | DOAJ |
| description | Existing anomaly detectors in AMIs suffer from shallow architectures, which impede their ability to capture temporal correlations and complex patterns in electricity consumption data, thus impact detection performance adversely. A deep (stacked) autoencoder structure based on Long Short-Term Memory (LSTM) with a sequence-to-sequence (seq2seq) configuration is proposed. The depth of the autoencoder architecture is beneficial for capturing complex data patterns, and the seq2seq LSTM model effectively utilizes the temporal sequential characteristics of the data. The performance of simple autoencoders, variational autoencoders, and Attention Enhanced Autoencoders (AEA) was studied, revealing that using the seq2seq structure in these three types of autoencoders results in superior detection performance compared to fully connected architectures. Simulation results demonstrate that the detector with an attention mechanism (AEA) achieves a 4%~21% higher detection rate and a 4%~13% lower false alarm rate compared to the best-performing existing detectors. |
| format | Article |
| id | doaj-art-d0c95c5ebf2c44118d0b1ea7f53f70bc |
| institution | Kabale University |
| issn | 0258-7998 |
| language | zho |
| publishDate | 2024-02-01 |
| publisher | National Computer System Engineering Research Institute of China |
| record_format | Article |
| series | Dianzi Jishu Yingyong |
| spelling | doaj-art-d0c95c5ebf2c44118d0b1ea7f53f70bc2025-08-20T03:30:35ZzhoNational Computer System Engineering Research Institute of ChinaDianzi Jishu Yingyong0258-79982024-02-01502768210.16157/j.issn.0258-7998.2343953000163482Anomaly detection of smart grid stealing network attacks based on deep autoencoderHuang Yan0Li Jincan1Yang Xiaqin2Li Pei3Li Zi4State Grid Guangxi Power Supply Company，Nanning 530023， ChinaState Grid Guangxi Power Supply Company，Nanning 530023， ChinaState Grid Nanning Power Supply Company，Nanning 530000， ChinaState Grid Nanning Power Supply Company，Nanning 530000， ChinaState Grid Wuzhou Power Supply Company，Wuzhou 543002， ChinaExisting anomaly detectors in AMIs suffer from shallow architectures, which impede their ability to capture temporal correlations and complex patterns in electricity consumption data, thus impact detection performance adversely. A deep (stacked) autoencoder structure based on Long Short-Term Memory (LSTM) with a sequence-to-sequence (seq2seq) configuration is proposed. The depth of the autoencoder architecture is beneficial for capturing complex data patterns, and the seq2seq LSTM model effectively utilizes the temporal sequential characteristics of the data. The performance of simple autoencoders, variational autoencoders, and Attention Enhanced Autoencoders (AEA) was studied, revealing that using the seq2seq structure in these three types of autoencoders results in superior detection performance compared to fully connected architectures. Simulation results demonstrate that the detector with an attention mechanism (AEA) achieves a 4%~21% higher detection rate and a 4%~13% lower false alarm rate compared to the best-performing existing detectors.http://www.chinaaet.com/article/3000163482autoencoderdeep machine learningpower stealinghyperparameter optimizationsequence-to-sequence |
| spellingShingle | Huang Yan Li Jincan Yang Xiaqin Li Pei Li Zi Anomaly detection of smart grid stealing network attacks based on deep autoencoder Dianzi Jishu Yingyong autoencoder deep machine learning power stealing hyperparameter optimization sequence-to-sequence |
| title | Anomaly detection of smart grid stealing network attacks based on deep autoencoder |
| title_full | Anomaly detection of smart grid stealing network attacks based on deep autoencoder |
| title_fullStr | Anomaly detection of smart grid stealing network attacks based on deep autoencoder |
| title_full_unstemmed | Anomaly detection of smart grid stealing network attacks based on deep autoencoder |
| title_short | Anomaly detection of smart grid stealing network attacks based on deep autoencoder |
| title_sort | anomaly detection of smart grid stealing network attacks based on deep autoencoder |
| topic | autoencoder deep machine learning power stealing hyperparameter optimization sequence-to-sequence |
| url | http://www.chinaaet.com/article/3000163482 |
| work_keys_str_mv | AT huangyan anomalydetectionofsmartgridstealingnetworkattacksbasedondeepautoencoder AT lijincan anomalydetectionofsmartgridstealingnetworkattacksbasedondeepautoencoder AT yangxiaqin anomalydetectionofsmartgridstealingnetworkattacksbasedondeepautoencoder AT lipei anomalydetectionofsmartgridstealingnetworkattacksbasedondeepautoencoder AT lizi anomalydetectionofsmartgridstealingnetworkattacksbasedondeepautoencoder |