Deep Learning-Based Non-Intrusive Commercial Load Monitoring
Commercial load is an essential demand-side resource. Monitoring commercial loads helps not only commercial customers understand their energy usage to improve energy efficiency but also helps electric utilities develop demand-side management strategies to ensure stable operation of the power system....
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-07-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/14/5250 |
| _version_ | 1797443866422411264 |
|---|---|
| author | Mengran Zhou Shuai Shao Xu Wang Ziwei Zhu Feng Hu |
| author_facet | Mengran Zhou Shuai Shao Xu Wang Ziwei Zhu Feng Hu |
| author_sort | Mengran Zhou |
| collection | DOAJ |
| description | Commercial load is an essential demand-side resource. Monitoring commercial loads helps not only commercial customers understand their energy usage to improve energy efficiency but also helps electric utilities develop demand-side management strategies to ensure stable operation of the power system. However, existing non-intrusive methods cannot monitor multiple commercial loads simultaneously and do not consider the high correlation and severe imbalance among commercial loads. Therefore, this paper proposes a deep learning-based non-intrusive commercial load monitoring method to solve these problems. The method takes the total power signal of the commercial building as input and directly determines the state and power consumption of several specific appliances. The key elements of the method are a new neural network structure called TTRNet and a new loss function called MLFL. TTRNet is a multi-label classification model that can autonomously learn correlation information through its unique network structure. MLFL is a loss function specifically designed for multi-label classification tasks, which solves the imbalance problem and improves the monitoring accuracy for challenging loads. To validate the proposed method, experiments are performed separately in seen and unseen scenarios using a public dataset. In the seen scenario, the method achieves an average F1 score of 0.957, which is 7.77% better than existing multi-label classification methods. In the unseen scenario, the average F1 score is 0.904, which is 1.92% better than existing methods. The experimental results show that the method proposed in this paper is both effective and practical. |
| first_indexed | 2024-03-09T13:04:15Z |
| format | Article |
| id | doaj.art-ed7f6f2130cf42c9ad162dda30bb6acf |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T13:04:15Z |
| publishDate | 2022-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-ed7f6f2130cf42c9ad162dda30bb6acf2023-11-30T21:51:30ZengMDPI AGSensors1424-82202022-07-012214525010.3390/s22145250Deep Learning-Based Non-Intrusive Commercial Load MonitoringMengran Zhou0Shuai Shao1Xu Wang2Ziwei Zhu3Feng Hu4School of Electrical and Information Engineering, Anhui University of Science and Technology, Huainan 232001, ChinaSchool of Electrical and Information Engineering, Anhui University of Science and Technology, Huainan 232001, ChinaSchool of Electrical and Information Engineering, Anhui University of Science and Technology, Huainan 232001, ChinaSchool of Electrical and Information Engineering, Anhui University of Science and Technology, Huainan 232001, ChinaSchool of Electrical and Information Engineering, Anhui University of Science and Technology, Huainan 232001, ChinaCommercial load is an essential demand-side resource. Monitoring commercial loads helps not only commercial customers understand their energy usage to improve energy efficiency but also helps electric utilities develop demand-side management strategies to ensure stable operation of the power system. However, existing non-intrusive methods cannot monitor multiple commercial loads simultaneously and do not consider the high correlation and severe imbalance among commercial loads. Therefore, this paper proposes a deep learning-based non-intrusive commercial load monitoring method to solve these problems. The method takes the total power signal of the commercial building as input and directly determines the state and power consumption of several specific appliances. The key elements of the method are a new neural network structure called TTRNet and a new loss function called MLFL. TTRNet is a multi-label classification model that can autonomously learn correlation information through its unique network structure. MLFL is a loss function specifically designed for multi-label classification tasks, which solves the imbalance problem and improves the monitoring accuracy for challenging loads. To validate the proposed method, experiments are performed separately in seen and unseen scenarios using a public dataset. In the seen scenario, the method achieves an average F1 score of 0.957, which is 7.77% better than existing multi-label classification methods. In the unseen scenario, the average F1 score is 0.904, which is 1.92% better than existing methods. The experimental results show that the method proposed in this paper is both effective and practical.https://www.mdpi.com/1424-8220/22/14/5250non-intrusive load monitoringcommercial loaddeep learningmulti-label classificationcorrelationimbalance |
| spellingShingle | Mengran Zhou Shuai Shao Xu Wang Ziwei Zhu Feng Hu Deep Learning-Based Non-Intrusive Commercial Load Monitoring Sensors non-intrusive load monitoring commercial load deep learning multi-label classification correlation imbalance |
| title | Deep Learning-Based Non-Intrusive Commercial Load Monitoring |
| title_full | Deep Learning-Based Non-Intrusive Commercial Load Monitoring |
| title_fullStr | Deep Learning-Based Non-Intrusive Commercial Load Monitoring |
| title_full_unstemmed | Deep Learning-Based Non-Intrusive Commercial Load Monitoring |
| title_short | Deep Learning-Based Non-Intrusive Commercial Load Monitoring |
| title_sort | deep learning based non intrusive commercial load monitoring |
| topic | non-intrusive load monitoring commercial load deep learning multi-label classification correlation imbalance |
| url | https://www.mdpi.com/1424-8220/22/14/5250 |
| work_keys_str_mv | AT mengranzhou deeplearningbasednonintrusivecommercialloadmonitoring AT shuaishao deeplearningbasednonintrusivecommercialloadmonitoring AT xuwang deeplearningbasednonintrusivecommercialloadmonitoring AT ziweizhu deeplearningbasednonintrusivecommercialloadmonitoring AT fenghu deeplearningbasednonintrusivecommercialloadmonitoring |