PLS-CNN-BiLSTM: An End-to-End Algorithm-Based Savitzky–Golay Smoothing and Evolution Strategy for Load Forecasting
This paper proposes an effective deep learning framework for Short-Term Load Forecasting (STLF) of multivariate time series. The proposed model consists of a hybrid Convolutional neural network-Bidirectional Long Short-Term Memory (CBiLSTM) based on the Evolution Strategy (ES) method and the Savitzk...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-10-01
|
| Series: | Energies |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1073/13/20/5464 |
| _version_ | 1827704029472358400 |
|---|---|
| author | Mohamed Massaoudi Shady S. Refaat Haitham Abu-Rub Ines Chihi Fakhreddine S. Oueslati |
| author_facet | Mohamed Massaoudi Shady S. Refaat Haitham Abu-Rub Ines Chihi Fakhreddine S. Oueslati |
| author_sort | Mohamed Massaoudi |
| collection | DOAJ |
| description | This paper proposes an effective deep learning framework for Short-Term Load Forecasting (STLF) of multivariate time series. The proposed model consists of a hybrid Convolutional neural network-Bidirectional Long Short-Term Memory (CBiLSTM) based on the Evolution Strategy (ES) method and the Savitzky–Golay (SG) filter (SG-CBiLSTM). The adopted methodology incorporates the virtue of different prepossessing blocks to enhance the performance of the CBiLSTM model. In particular, a data-augmentation strategy is employed to synthetically improve the feature representation of the CBiLSTM model. The augmented data is forwarded to the Partial Least Square (PLS) method to select the most informative features above the predefined threshold. Next, the SG algorithm is computed for smoothing the load to enhance the learning capabilities of the underlying system. The structure of the SG-CBiLSTM for the ISO New England dataset is optimized using the ES technique. Finally, the CBiLSTM model generates output forecasts. The proposed approach demonstrates a remarkable improvement in the performance of the original CBiLSTM model. Furthermore, the experimental results strongly confirm the high effectiveness of the proposed SG-CBiLSTM model compared to the state-of-the-art techniques. |
| first_indexed | 2024-03-10T15:31:03Z |
| format | Article |
| id | doaj.art-ecf058e5a3144823b247354d8bc8989b |
| institution | Directory Open Access Journal |
| issn | 1996-1073 |
| language | English |
| last_indexed | 2024-03-10T15:31:03Z |
| publishDate | 2020-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Energies |
| spelling | doaj.art-ecf058e5a3144823b247354d8bc8989b2023-11-20T17:41:33ZengMDPI AGEnergies1996-10732020-10-011320546410.3390/en13205464PLS-CNN-BiLSTM: An End-to-End Algorithm-Based Savitzky–Golay Smoothing and Evolution Strategy for Load ForecastingMohamed Massaoudi0Shady S. Refaat1Haitham Abu-Rub2Ines Chihi3Fakhreddine S. Oueslati4Department of Electrical and Computer Engineering, Texas A and M University at Qatar, Doha 3263, QatarDepartment of Electrical and Computer Engineering, Texas A and M University at Qatar, Doha 3263, QatarDepartment of Electrical and Computer Engineering, Texas A and M University at Qatar, Doha 3263, QatarLaboratory of Energy Applications and Renewable Energy Efficiency (LAPER), El Manar University, Tunis 1002, TunisiaUnité de Recherche de Physique des Semi-Conducteurs et Capteurs, Carthage University, Tunis 2070, TunisiaThis paper proposes an effective deep learning framework for Short-Term Load Forecasting (STLF) of multivariate time series. The proposed model consists of a hybrid Convolutional neural network-Bidirectional Long Short-Term Memory (CBiLSTM) based on the Evolution Strategy (ES) method and the Savitzky–Golay (SG) filter (SG-CBiLSTM). The adopted methodology incorporates the virtue of different prepossessing blocks to enhance the performance of the CBiLSTM model. In particular, a data-augmentation strategy is employed to synthetically improve the feature representation of the CBiLSTM model. The augmented data is forwarded to the Partial Least Square (PLS) method to select the most informative features above the predefined threshold. Next, the SG algorithm is computed for smoothing the load to enhance the learning capabilities of the underlying system. The structure of the SG-CBiLSTM for the ISO New England dataset is optimized using the ES technique. Finally, the CBiLSTM model generates output forecasts. The proposed approach demonstrates a remarkable improvement in the performance of the original CBiLSTM model. Furthermore, the experimental results strongly confirm the high effectiveness of the proposed SG-CBiLSTM model compared to the state-of-the-art techniques.https://www.mdpi.com/1996-1073/13/20/5464Bidirectional Long Short-Term Memory (BiLSTM)Convolutional Neural Network (CNN)evolution strategyPartial Least Square (PLS) methodSavitzky–GolayShort-Term Load Forecasting (STLF) |
| spellingShingle | Mohamed Massaoudi Shady S. Refaat Haitham Abu-Rub Ines Chihi Fakhreddine S. Oueslati PLS-CNN-BiLSTM: An End-to-End Algorithm-Based Savitzky–Golay Smoothing and Evolution Strategy for Load Forecasting Energies Bidirectional Long Short-Term Memory (BiLSTM) Convolutional Neural Network (CNN) evolution strategy Partial Least Square (PLS) method Savitzky–Golay Short-Term Load Forecasting (STLF) |
| title | PLS-CNN-BiLSTM: An End-to-End Algorithm-Based Savitzky–Golay Smoothing and Evolution Strategy for Load Forecasting |
| title_full | PLS-CNN-BiLSTM: An End-to-End Algorithm-Based Savitzky–Golay Smoothing and Evolution Strategy for Load Forecasting |
| title_fullStr | PLS-CNN-BiLSTM: An End-to-End Algorithm-Based Savitzky–Golay Smoothing and Evolution Strategy for Load Forecasting |
| title_full_unstemmed | PLS-CNN-BiLSTM: An End-to-End Algorithm-Based Savitzky–Golay Smoothing and Evolution Strategy for Load Forecasting |
| title_short | PLS-CNN-BiLSTM: An End-to-End Algorithm-Based Savitzky–Golay Smoothing and Evolution Strategy for Load Forecasting |
| title_sort | pls cnn bilstm an end to end algorithm based savitzky golay smoothing and evolution strategy for load forecasting |
| topic | Bidirectional Long Short-Term Memory (BiLSTM) Convolutional Neural Network (CNN) evolution strategy Partial Least Square (PLS) method Savitzky–Golay Short-Term Load Forecasting (STLF) |
| url | https://www.mdpi.com/1996-1073/13/20/5464 |
| work_keys_str_mv | AT mohamedmassaoudi plscnnbilstmanendtoendalgorithmbasedsavitzkygolaysmoothingandevolutionstrategyforloadforecasting AT shadysrefaat plscnnbilstmanendtoendalgorithmbasedsavitzkygolaysmoothingandevolutionstrategyforloadforecasting AT haithamaburub plscnnbilstmanendtoendalgorithmbasedsavitzkygolaysmoothingandevolutionstrategyforloadforecasting AT ineschihi plscnnbilstmanendtoendalgorithmbasedsavitzkygolaysmoothingandevolutionstrategyforloadforecasting AT fakhreddinesoueslati plscnnbilstmanendtoendalgorithmbasedsavitzkygolaysmoothingandevolutionstrategyforloadforecasting |