Learning Coupled Meteorological Characteristics Aids Short-Term Photovoltaic Interval Prediction Methods
In response to the challenges posed by renewable energy integration, this study introduces a hybrid Attention-TCN-LSTM model for short-term photovoltaic (PV) power forecasting. The LSTM captures the sequence characteristics of PV output, which are then combined with the meteorological sequence featu...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Energies |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1073/18/2/308 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588528278568960 |
|---|---|
| author | Yue Guo Yu Song Zilong Lai Xuyang Wang Licheng Wang Hui Qin |
| author_facet | Yue Guo Yu Song Zilong Lai Xuyang Wang Licheng Wang Hui Qin |
| author_sort | Yue Guo |
| collection | DOAJ |
| description | In response to the challenges posed by renewable energy integration, this study introduces a hybrid Attention-TCN-LSTM model for short-term photovoltaic (PV) power forecasting. The LSTM captures the sequence characteristics of PV output, which are then combined with the meteorological sequence features extracted by the Attention-TCN module. The model leverages the strengths of the TCN, the LSTM, and the self-attention mechanism to enhance prediction accuracy and construct reliable prediction intervals. Aiming to optimize both performance and efficiency, the PSO algorithm is used for hyperparameter optimization. Ablation studies and comparisons with other models confirm the effectiveness, accuracy and robustness of the proposed model. This hybrid approach contributes to improved renewable energy integration, offering a more stable and reliable energy supply. Future work will focus on incorporating intelligent systems for autonomous risk management and real-time control of dynamic PV output fluctuations. |
| format | Article |
| id | doaj-art-047cbcd1232d486cb1cbe43b7e54310a |
| institution | Kabale University |
| issn | 1996-1073 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Energies |
| spelling | doaj-art-047cbcd1232d486cb1cbe43b7e54310a2025-01-24T13:30:58ZengMDPI AGEnergies1996-10732025-01-0118230810.3390/en18020308Learning Coupled Meteorological Characteristics Aids Short-Term Photovoltaic Interval Prediction MethodsYue Guo0Yu Song1Zilong Lai2Xuyang Wang3Licheng Wang4Hui Qin5State Grid Economic and Technological Research Institute, Beijing 102209, ChinaState Grid Economic and Technological Research Institute, Beijing 102209, ChinaCollege of Information Engineering, Zhejiang University of Technology, Hangzhou 310027, ChinaState Grid Economic and Technological Research Institute, Beijing 102209, ChinaCollege of Information Engineering, Zhejiang University of Technology, Hangzhou 310027, ChinaCollege of Information Engineering, Zhejiang University of Technology, Hangzhou 310027, ChinaIn response to the challenges posed by renewable energy integration, this study introduces a hybrid Attention-TCN-LSTM model for short-term photovoltaic (PV) power forecasting. The LSTM captures the sequence characteristics of PV output, which are then combined with the meteorological sequence features extracted by the Attention-TCN module. The model leverages the strengths of the TCN, the LSTM, and the self-attention mechanism to enhance prediction accuracy and construct reliable prediction intervals. Aiming to optimize both performance and efficiency, the PSO algorithm is used for hyperparameter optimization. Ablation studies and comparisons with other models confirm the effectiveness, accuracy and robustness of the proposed model. This hybrid approach contributes to improved renewable energy integration, offering a more stable and reliable energy supply. Future work will focus on incorporating intelligent systems for autonomous risk management and real-time control of dynamic PV output fluctuations.https://www.mdpi.com/1996-1073/18/2/308PV forecastLSTMtemporal convolutional network (TCN) |
| spellingShingle | Yue Guo Yu Song Zilong Lai Xuyang Wang Licheng Wang Hui Qin Learning Coupled Meteorological Characteristics Aids Short-Term Photovoltaic Interval Prediction Methods Energies PV forecast LSTM temporal convolutional network (TCN) |
| title | Learning Coupled Meteorological Characteristics Aids Short-Term Photovoltaic Interval Prediction Methods |
| title_full | Learning Coupled Meteorological Characteristics Aids Short-Term Photovoltaic Interval Prediction Methods |
| title_fullStr | Learning Coupled Meteorological Characteristics Aids Short-Term Photovoltaic Interval Prediction Methods |
| title_full_unstemmed | Learning Coupled Meteorological Characteristics Aids Short-Term Photovoltaic Interval Prediction Methods |
| title_short | Learning Coupled Meteorological Characteristics Aids Short-Term Photovoltaic Interval Prediction Methods |
| title_sort | learning coupled meteorological characteristics aids short term photovoltaic interval prediction methods |
| topic | PV forecast LSTM temporal convolutional network (TCN) |
| url | https://www.mdpi.com/1996-1073/18/2/308 |
| work_keys_str_mv | AT yueguo learningcoupledmeteorologicalcharacteristicsaidsshorttermphotovoltaicintervalpredictionmethods AT yusong learningcoupledmeteorologicalcharacteristicsaidsshorttermphotovoltaicintervalpredictionmethods AT zilonglai learningcoupledmeteorologicalcharacteristicsaidsshorttermphotovoltaicintervalpredictionmethods AT xuyangwang learningcoupledmeteorologicalcharacteristicsaidsshorttermphotovoltaicintervalpredictionmethods AT lichengwang learningcoupledmeteorologicalcharacteristicsaidsshorttermphotovoltaicintervalpredictionmethods AT huiqin learningcoupledmeteorologicalcharacteristicsaidsshorttermphotovoltaicintervalpredictionmethods |