Testing the Wind Energy Data Based on Environmental Factors Predicted by Machine Learning with Analysis of Variance
This study proposes a two-stage methodology for predicting wind energy production using time, environmental, technical, and locational variables. In the first stage, machine learning algorithms, including random forest (RF), gradient boosting (GB), k-nearest neighbors (kNNs), linear regression (LR),...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/15/1/241 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841549420454215680 |
|---|---|
| author | Yasemin Ayaz Atalan Abdulkadir Atalan |
| author_facet | Yasemin Ayaz Atalan Abdulkadir Atalan |
| author_sort | Yasemin Ayaz Atalan |
| collection | DOAJ |
| description | This study proposes a two-stage methodology for predicting wind energy production using time, environmental, technical, and locational variables. In the first stage, machine learning algorithms, including random forest (RF), gradient boosting (GB), k-nearest neighbors (kNNs), linear regression (LR), and decision trees (Tree), were employed to estimate energy output. Among these, RF exhibited the best performance with the lowest error metrics (MSE: 0.003, RMSE: 0.053) and the highest R<sup>2</sup> value (0.988). In the second stage, analysis of variance (ANOVA) was conducted to evaluate the statistical relationships between independent variables and the predicted dependent variable, identifying wind speed (<i>p</i> < 0.001) and rotor speed (<i>p</i> < 0.001) as the most influential factors. Furthermore, RF and GB models produced predictions most closely aligned with actual data, achieving R<sup>2</sup> values of 88.83% and 89.30% in the ANOVA validation phase. Integrating RF and GB models with statistical validation highlighted the robustness of the methodology. These findings demonstrate the robustness of integrating machine learning models with statistical verification methods. |
| format | Article |
| id | doaj-art-e05b4d2ecdae41bab650db7f50b4b404 |
| institution | Kabale University |
| issn | 2076-3417 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj-art-e05b4d2ecdae41bab650db7f50b4b4042025-01-10T13:14:54ZengMDPI AGApplied Sciences2076-34172024-12-0115124110.3390/app15010241Testing the Wind Energy Data Based on Environmental Factors Predicted by Machine Learning with Analysis of VarianceYasemin Ayaz Atalan0Abdulkadir Atalan1Department of Energy Management, Çanakkale Onsekiz Mart University, Çanakkale 17100, TurkeyFaculty of Engineering, Çanakkale Onsekiz Mart University, Çanakkale 17100, TurkeyThis study proposes a two-stage methodology for predicting wind energy production using time, environmental, technical, and locational variables. In the first stage, machine learning algorithms, including random forest (RF), gradient boosting (GB), k-nearest neighbors (kNNs), linear regression (LR), and decision trees (Tree), were employed to estimate energy output. Among these, RF exhibited the best performance with the lowest error metrics (MSE: 0.003, RMSE: 0.053) and the highest R<sup>2</sup> value (0.988). In the second stage, analysis of variance (ANOVA) was conducted to evaluate the statistical relationships between independent variables and the predicted dependent variable, identifying wind speed (<i>p</i> < 0.001) and rotor speed (<i>p</i> < 0.001) as the most influential factors. Furthermore, RF and GB models produced predictions most closely aligned with actual data, achieving R<sup>2</sup> values of 88.83% and 89.30% in the ANOVA validation phase. Integrating RF and GB models with statistical validation highlighted the robustness of the methodology. These findings demonstrate the robustness of integrating machine learning models with statistical verification methods.https://www.mdpi.com/2076-3417/15/1/241wind energy predictionrenewable energymachine learningANOVAstatistical validation |
| spellingShingle | Yasemin Ayaz Atalan Abdulkadir Atalan Testing the Wind Energy Data Based on Environmental Factors Predicted by Machine Learning with Analysis of Variance Applied Sciences wind energy prediction renewable energy machine learning ANOVA statistical validation |
| title | Testing the Wind Energy Data Based on Environmental Factors Predicted by Machine Learning with Analysis of Variance |
| title_full | Testing the Wind Energy Data Based on Environmental Factors Predicted by Machine Learning with Analysis of Variance |
| title_fullStr | Testing the Wind Energy Data Based on Environmental Factors Predicted by Machine Learning with Analysis of Variance |
| title_full_unstemmed | Testing the Wind Energy Data Based on Environmental Factors Predicted by Machine Learning with Analysis of Variance |
| title_short | Testing the Wind Energy Data Based on Environmental Factors Predicted by Machine Learning with Analysis of Variance |
| title_sort | testing the wind energy data based on environmental factors predicted by machine learning with analysis of variance |
| topic | wind energy prediction renewable energy machine learning ANOVA statistical validation |
| url | https://www.mdpi.com/2076-3417/15/1/241 |
| work_keys_str_mv | AT yaseminayazatalan testingthewindenergydatabasedonenvironmentalfactorspredictedbymachinelearningwithanalysisofvariance AT abdulkadiratalan testingthewindenergydatabasedonenvironmentalfactorspredictedbymachinelearningwithanalysisofvariance |