A Long Short-Term Memory–Wasserstein Generative Adversarial Network-Based Data Imputation Method for Photovoltaic Power Output Prediction
To address the challenges of the issue of inaccurate prediction results due to missing data in PV power records, a photovoltaic power data imputation method based on a Wasserstein Generative Adversarial Network (WGAN) and Long Short-Term Memory (LSTM) network is proposed. This method introduces a da...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Energies |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1073/18/2/399 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588538277789696 |
|---|---|
| author | Zhu Liu Lingfeng Xuan Dehuang Gong Xinlin Xie Dongguo Zhou |
| author_facet | Zhu Liu Lingfeng Xuan Dehuang Gong Xinlin Xie Dongguo Zhou |
| author_sort | Zhu Liu |
| collection | DOAJ |
| description | To address the challenges of the issue of inaccurate prediction results due to missing data in PV power records, a photovoltaic power data imputation method based on a Wasserstein Generative Adversarial Network (WGAN) and Long Short-Term Memory (LSTM) network is proposed. This method introduces a data-driven GAN framework with quasi-convex characteristics to ensure the smoothness of the imputed data with the existing data and employs a gradient penalty mechanism and a single-batch multi-iteration strategy for stable training. Finally, through frequency domain analysis, t-Distributed Stochastic Neighbor Embedding (t-SNE) metrics, and prediction performance validation of the generated data, the proposed method can improve the continuity and reliability of data in photovoltaic prediction tasks. |
| format | Article |
| id | doaj-art-2eb528bc793f4571a276ae7ceefa2d5d |
| institution | Kabale University |
| issn | 1996-1073 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Energies |
| spelling | doaj-art-2eb528bc793f4571a276ae7ceefa2d5d2025-01-24T13:31:21ZengMDPI AGEnergies1996-10732025-01-0118239910.3390/en18020399A Long Short-Term Memory–Wasserstein Generative Adversarial Network-Based Data Imputation Method for Photovoltaic Power Output PredictionZhu Liu0Lingfeng Xuan1Dehuang Gong2Xinlin Xie3Dongguo Zhou4China Southern Power Grid Research Technology Co., Ltd., Guangzhou 510663, ChinaQingyuan Yingde Power Supply Bureau, Guangdong Electric Power Co., Ltd., Yingde 513000, ChinaQingyuan Yingde Power Supply Bureau, Guangdong Electric Power Co., Ltd., Yingde 513000, ChinaQingyuan Yingde Power Supply Bureau, Guangdong Electric Power Co., Ltd., Yingde 513000, ChinaSchool of Electrical Engineering and Automation, Wuhan University, Wuhan 430072, ChinaTo address the challenges of the issue of inaccurate prediction results due to missing data in PV power records, a photovoltaic power data imputation method based on a Wasserstein Generative Adversarial Network (WGAN) and Long Short-Term Memory (LSTM) network is proposed. This method introduces a data-driven GAN framework with quasi-convex characteristics to ensure the smoothness of the imputed data with the existing data and employs a gradient penalty mechanism and a single-batch multi-iteration strategy for stable training. Finally, through frequency domain analysis, t-Distributed Stochastic Neighbor Embedding (t-SNE) metrics, and prediction performance validation of the generated data, the proposed method can improve the continuity and reliability of data in photovoltaic prediction tasks.https://www.mdpi.com/1996-1073/18/2/399PV output predictiondata imputationGANLSTMgradient penalty mechanism |
| spellingShingle | Zhu Liu Lingfeng Xuan Dehuang Gong Xinlin Xie Dongguo Zhou A Long Short-Term Memory–Wasserstein Generative Adversarial Network-Based Data Imputation Method for Photovoltaic Power Output Prediction Energies PV output prediction data imputation GAN LSTM gradient penalty mechanism |
| title | A Long Short-Term Memory–Wasserstein Generative Adversarial Network-Based Data Imputation Method for Photovoltaic Power Output Prediction |
| title_full | A Long Short-Term Memory–Wasserstein Generative Adversarial Network-Based Data Imputation Method for Photovoltaic Power Output Prediction |
| title_fullStr | A Long Short-Term Memory–Wasserstein Generative Adversarial Network-Based Data Imputation Method for Photovoltaic Power Output Prediction |
| title_full_unstemmed | A Long Short-Term Memory–Wasserstein Generative Adversarial Network-Based Data Imputation Method for Photovoltaic Power Output Prediction |
| title_short | A Long Short-Term Memory–Wasserstein Generative Adversarial Network-Based Data Imputation Method for Photovoltaic Power Output Prediction |
| title_sort | long short term memory wasserstein generative adversarial network based data imputation method for photovoltaic power output prediction |
| topic | PV output prediction data imputation GAN LSTM gradient penalty mechanism |
| url | https://www.mdpi.com/1996-1073/18/2/399 |
| work_keys_str_mv | AT zhuliu alongshorttermmemorywassersteingenerativeadversarialnetworkbaseddataimputationmethodforphotovoltaicpoweroutputprediction AT lingfengxuan alongshorttermmemorywassersteingenerativeadversarialnetworkbaseddataimputationmethodforphotovoltaicpoweroutputprediction AT dehuanggong alongshorttermmemorywassersteingenerativeadversarialnetworkbaseddataimputationmethodforphotovoltaicpoweroutputprediction AT xinlinxie alongshorttermmemorywassersteingenerativeadversarialnetworkbaseddataimputationmethodforphotovoltaicpoweroutputprediction AT dongguozhou alongshorttermmemorywassersteingenerativeadversarialnetworkbaseddataimputationmethodforphotovoltaicpoweroutputprediction AT zhuliu longshorttermmemorywassersteingenerativeadversarialnetworkbaseddataimputationmethodforphotovoltaicpoweroutputprediction AT lingfengxuan longshorttermmemorywassersteingenerativeadversarialnetworkbaseddataimputationmethodforphotovoltaicpoweroutputprediction AT dehuanggong longshorttermmemorywassersteingenerativeadversarialnetworkbaseddataimputationmethodforphotovoltaicpoweroutputprediction AT xinlinxie longshorttermmemorywassersteingenerativeadversarialnetworkbaseddataimputationmethodforphotovoltaicpoweroutputprediction AT dongguozhou longshorttermmemorywassersteingenerativeadversarialnetworkbaseddataimputationmethodforphotovoltaicpoweroutputprediction |