Ultra-Short-Term Distributed Photovoltaic Power Probabilistic Forecasting Method Based on Federated Learning and Joint Probability Distribution Modeling
The accurate probabilistic forecasting of ultra-short-term power generation from distributed photovoltaic (DPV) systems is of great significance for optimizing electricity markets and managing energy on the user side. Existing methods regarding cluster information sharing tend to easily trigger issu...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Energies |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1073/18/1/197 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841549214511792128 |
|---|---|
| author | Yubo Wang Chao Huo Fei Xu Libin Zheng Ling Hao |
| author_facet | Yubo Wang Chao Huo Fei Xu Libin Zheng Ling Hao |
| author_sort | Yubo Wang |
| collection | DOAJ |
| description | The accurate probabilistic forecasting of ultra-short-term power generation from distributed photovoltaic (DPV) systems is of great significance for optimizing electricity markets and managing energy on the user side. Existing methods regarding cluster information sharing tend to easily trigger issues of data privacy leakage during information sharing, or they suffer from insufficient information sharing while protecting data privacy, leading to suboptimal forecasting performance. To address these issues, this paper proposes a privacy-preserving deep federated learning method for the probabilistic forecasting of ultra-short-term power generation from DPV systems. Firstly, a collaborative feature federated learning framework is established. For the central server, information sharing among clients is realized through the interaction of global models and features while avoiding the direct interaction of raw data to ensure the security of client data privacy. For local clients, a Transformer autoencoder is used as the forecasting model to extract local temporal features, which are combined with global features to form spatiotemporal correlation features, thereby deeply exploring the spatiotemporal correlations between different power stations and improving the accuracy of forecasting. Subsequently, a joint probability distribution model of forecasting values and errors is constructed, and the distribution patterns of errors are finely studied based on the dependencies between data to enhance the accuracy of probabilistic forecasting. Finally, the effectiveness of the proposed method was validated through real datasets. |
| format | Article |
| id | doaj-art-5f4df88b2f9b439695070ec439b6f475 |
| institution | Kabale University |
| issn | 1996-1073 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Energies |
| spelling | doaj-art-5f4df88b2f9b439695070ec439b6f4752025-01-10T13:17:23ZengMDPI AGEnergies1996-10732025-01-0118119710.3390/en18010197Ultra-Short-Term Distributed Photovoltaic Power Probabilistic Forecasting Method Based on Federated Learning and Joint Probability Distribution ModelingYubo Wang0Chao Huo1Fei Xu2Libin Zheng3Ling Hao4Beijing SmartChip Microelectronics Technology Company Limited, Beijing 102200, ChinaBeijing SmartChip Microelectronics Technology Company Limited, Beijing 102200, ChinaState Key Laboratory of Power System Operation and Control (Department of Electrical Engineering), Tsinghua University, Beijing 100084, ChinaBeijing SmartChip Microelectronics Technology Company Limited, Beijing 102200, ChinaState Key Laboratory of Power System Operation and Control (Department of Electrical Engineering), Tsinghua University, Beijing 100084, ChinaThe accurate probabilistic forecasting of ultra-short-term power generation from distributed photovoltaic (DPV) systems is of great significance for optimizing electricity markets and managing energy on the user side. Existing methods regarding cluster information sharing tend to easily trigger issues of data privacy leakage during information sharing, or they suffer from insufficient information sharing while protecting data privacy, leading to suboptimal forecasting performance. To address these issues, this paper proposes a privacy-preserving deep federated learning method for the probabilistic forecasting of ultra-short-term power generation from DPV systems. Firstly, a collaborative feature federated learning framework is established. For the central server, information sharing among clients is realized through the interaction of global models and features while avoiding the direct interaction of raw data to ensure the security of client data privacy. For local clients, a Transformer autoencoder is used as the forecasting model to extract local temporal features, which are combined with global features to form spatiotemporal correlation features, thereby deeply exploring the spatiotemporal correlations between different power stations and improving the accuracy of forecasting. Subsequently, a joint probability distribution model of forecasting values and errors is constructed, and the distribution patterns of errors are finely studied based on the dependencies between data to enhance the accuracy of probabilistic forecasting. Finally, the effectiveness of the proposed method was validated through real datasets.https://www.mdpi.com/1996-1073/18/1/197distributed photovoltaicultra-short-term power forecastingfederated learningspatiotemporal correlationjoint probability distribution |
| spellingShingle | Yubo Wang Chao Huo Fei Xu Libin Zheng Ling Hao Ultra-Short-Term Distributed Photovoltaic Power Probabilistic Forecasting Method Based on Federated Learning and Joint Probability Distribution Modeling Energies distributed photovoltaic ultra-short-term power forecasting federated learning spatiotemporal correlation joint probability distribution |
| title | Ultra-Short-Term Distributed Photovoltaic Power Probabilistic Forecasting Method Based on Federated Learning and Joint Probability Distribution Modeling |
| title_full | Ultra-Short-Term Distributed Photovoltaic Power Probabilistic Forecasting Method Based on Federated Learning and Joint Probability Distribution Modeling |
| title_fullStr | Ultra-Short-Term Distributed Photovoltaic Power Probabilistic Forecasting Method Based on Federated Learning and Joint Probability Distribution Modeling |
| title_full_unstemmed | Ultra-Short-Term Distributed Photovoltaic Power Probabilistic Forecasting Method Based on Federated Learning and Joint Probability Distribution Modeling |
| title_short | Ultra-Short-Term Distributed Photovoltaic Power Probabilistic Forecasting Method Based on Federated Learning and Joint Probability Distribution Modeling |
| title_sort | ultra short term distributed photovoltaic power probabilistic forecasting method based on federated learning and joint probability distribution modeling |
| topic | distributed photovoltaic ultra-short-term power forecasting federated learning spatiotemporal correlation joint probability distribution |
| url | https://www.mdpi.com/1996-1073/18/1/197 |
| work_keys_str_mv | AT yubowang ultrashorttermdistributedphotovoltaicpowerprobabilisticforecastingmethodbasedonfederatedlearningandjointprobabilitydistributionmodeling AT chaohuo ultrashorttermdistributedphotovoltaicpowerprobabilisticforecastingmethodbasedonfederatedlearningandjointprobabilitydistributionmodeling AT feixu ultrashorttermdistributedphotovoltaicpowerprobabilisticforecastingmethodbasedonfederatedlearningandjointprobabilitydistributionmodeling AT libinzheng ultrashorttermdistributedphotovoltaicpowerprobabilisticforecastingmethodbasedonfederatedlearningandjointprobabilitydistributionmodeling AT linghao ultrashorttermdistributedphotovoltaicpowerprobabilisticforecastingmethodbasedonfederatedlearningandjointprobabilitydistributionmodeling |