Comparative analysis of deep neural network architectures for renewable energy forecasting: enhancing accuracy with meteorological and time-based features
Abstract This study evaluates and differentiates five advanced machine learning models—LSTM, GRU, CNN-LSTM, Random Forest, and SVR—aimed at precisely estimating solar and wind power generation to enhance renewable energy forecasting. LSTM achieved a remarkable Mean Squared Error (MSE) of 0.010 and R...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2024-12-01
|
| Series: | Discover Sustainability |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s43621-024-00783-5 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841559905377452032 |
|---|---|
| author | Sunawar Khan Tehseen Mazhar Muhammad Amir Khan Tariq Shahzad Wasim Ahmad Afsha Bibi Mamoon M. Saeed Habib Hamam |
| author_facet | Sunawar Khan Tehseen Mazhar Muhammad Amir Khan Tariq Shahzad Wasim Ahmad Afsha Bibi Mamoon M. Saeed Habib Hamam |
| author_sort | Sunawar Khan |
| collection | DOAJ |
| description | Abstract This study evaluates and differentiates five advanced machine learning models—LSTM, GRU, CNN-LSTM, Random Forest, and SVR—aimed at precisely estimating solar and wind power generation to enhance renewable energy forecasting. LSTM achieved a remarkable Mean Squared Error (MSE) of 0.010 and R2 score of 0.90, highlighting its proficiency in capturing intricate temporal relationships. GRU closely followed, demonstrating potential as a viable option due to its remarkable combination of computational efficiency and accuracy (MSE = 0.015, R2 = 0.88). In datasets abundant in spatial correlations, the CNN-LSTM hybrid demonstrated its utility by providing novel insights into spatial–temporal patterns; nonetheless, it lagged considerably in accuracy, with a mean squared error (MSE) of 0.020 and a R2 of 0.87. Conversely, traditional models demonstrated a reliable albeit less dynamic ability to elucidate the complexities of renewable energy data; for instance, Random Forest exhibited a mean squared error (MSE) of 0.025, while Support Vector Regression (SVR) recorded an MSE of 0.030. The results affirm that deep learning architectures, particularly LSTM, offer a transformative method for renewable energy forecasting, hence enhancing accuracy and reliability in energy management systems. |
| format | Article |
| id | doaj-art-b02168fc24d943a8be1770bbc67afed1 |
| institution | Kabale University |
| issn | 2662-9984 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Springer |
| record_format | Article |
| series | Discover Sustainability |
| spelling | doaj-art-b02168fc24d943a8be1770bbc67afed12025-01-05T12:05:16ZengSpringerDiscover Sustainability2662-99842024-12-015112410.1007/s43621-024-00783-5Comparative analysis of deep neural network architectures for renewable energy forecasting: enhancing accuracy with meteorological and time-based featuresSunawar Khan0Tehseen Mazhar1Muhammad Amir Khan2Tariq Shahzad3Wasim Ahmad4Afsha Bibi5Mamoon M. Saeed6Habib Hamam7Department of Software Engineering, Islamia University of BahawalpurSchool of Computer Science, National College of Business Administration and EconomicsSchool of Computing Sciences, College of Computing, Informatics, and Mathematics, Universiti Teknologi MARADepartment of Computer Science, COMSATS University Islamabad, Sahiwal CampusDepartment of Computing, School of Arts and Creative Technology, University of Greater ManchesterDepartment of Computer and Software Technology, University of SwatDepartment of Communications and Electronics Engineering, Faculty of Engineering, University of Modern Sciences (UMS)Faculty of Engineering, Uni de MonctonAbstract This study evaluates and differentiates five advanced machine learning models—LSTM, GRU, CNN-LSTM, Random Forest, and SVR—aimed at precisely estimating solar and wind power generation to enhance renewable energy forecasting. LSTM achieved a remarkable Mean Squared Error (MSE) of 0.010 and R2 score of 0.90, highlighting its proficiency in capturing intricate temporal relationships. GRU closely followed, demonstrating potential as a viable option due to its remarkable combination of computational efficiency and accuracy (MSE = 0.015, R2 = 0.88). In datasets abundant in spatial correlations, the CNN-LSTM hybrid demonstrated its utility by providing novel insights into spatial–temporal patterns; nonetheless, it lagged considerably in accuracy, with a mean squared error (MSE) of 0.020 and a R2 of 0.87. Conversely, traditional models demonstrated a reliable albeit less dynamic ability to elucidate the complexities of renewable energy data; for instance, Random Forest exhibited a mean squared error (MSE) of 0.025, while Support Vector Regression (SVR) recorded an MSE of 0.030. The results affirm that deep learning architectures, particularly LSTM, offer a transformative method for renewable energy forecasting, hence enhancing accuracy and reliability in energy management systems.https://doi.org/10.1007/s43621-024-00783-5Renewable energyForecastingDeep learningLSTMNeural networksHyperparameter tuning |
| spellingShingle | Sunawar Khan Tehseen Mazhar Muhammad Amir Khan Tariq Shahzad Wasim Ahmad Afsha Bibi Mamoon M. Saeed Habib Hamam Comparative analysis of deep neural network architectures for renewable energy forecasting: enhancing accuracy with meteorological and time-based features Discover Sustainability Renewable energy Forecasting Deep learning LSTM Neural networks Hyperparameter tuning |
| title | Comparative analysis of deep neural network architectures for renewable energy forecasting: enhancing accuracy with meteorological and time-based features |
| title_full | Comparative analysis of deep neural network architectures for renewable energy forecasting: enhancing accuracy with meteorological and time-based features |
| title_fullStr | Comparative analysis of deep neural network architectures for renewable energy forecasting: enhancing accuracy with meteorological and time-based features |
| title_full_unstemmed | Comparative analysis of deep neural network architectures for renewable energy forecasting: enhancing accuracy with meteorological and time-based features |
| title_short | Comparative analysis of deep neural network architectures for renewable energy forecasting: enhancing accuracy with meteorological and time-based features |
| title_sort | comparative analysis of deep neural network architectures for renewable energy forecasting enhancing accuracy with meteorological and time based features |
| topic | Renewable energy Forecasting Deep learning LSTM Neural networks Hyperparameter tuning |
| url | https://doi.org/10.1007/s43621-024-00783-5 |
| work_keys_str_mv | AT sunawarkhan comparativeanalysisofdeepneuralnetworkarchitecturesforrenewableenergyforecastingenhancingaccuracywithmeteorologicalandtimebasedfeatures AT tehseenmazhar comparativeanalysisofdeepneuralnetworkarchitecturesforrenewableenergyforecastingenhancingaccuracywithmeteorologicalandtimebasedfeatures AT muhammadamirkhan comparativeanalysisofdeepneuralnetworkarchitecturesforrenewableenergyforecastingenhancingaccuracywithmeteorologicalandtimebasedfeatures AT tariqshahzad comparativeanalysisofdeepneuralnetworkarchitecturesforrenewableenergyforecastingenhancingaccuracywithmeteorologicalandtimebasedfeatures AT wasimahmad comparativeanalysisofdeepneuralnetworkarchitecturesforrenewableenergyforecastingenhancingaccuracywithmeteorologicalandtimebasedfeatures AT afshabibi comparativeanalysisofdeepneuralnetworkarchitecturesforrenewableenergyforecastingenhancingaccuracywithmeteorologicalandtimebasedfeatures AT mamoonmsaeed comparativeanalysisofdeepneuralnetworkarchitecturesforrenewableenergyforecastingenhancingaccuracywithmeteorologicalandtimebasedfeatures AT habibhamam comparativeanalysisofdeepneuralnetworkarchitecturesforrenewableenergyforecastingenhancingaccuracywithmeteorologicalandtimebasedfeatures |