Short-term forecasting of global horizontal irradiance using stacked ensemble machine learning alogorithms
| dc.contributor.advisor | Ravele, T. | |
| dc.contributor.advisor | Sigauke, C. | |
| dc.contributor.author | Mugware, Fhulufhelo Walter | |
| dc.date | 2025 | |
| dc.date.accessioned | 2026-01-14T14:25:00Z | |
| dc.date.available | 2026-01-14T14:25:00Z | |
| dc.date.issued | 2025-09-05 | |
| dc.description | M. Sc (E-Science) | |
| dc.description.abstract | In today’s world, where sustainable energy is essential for the planet’s survival, accurate solar energy forecasting is crucial. This study focused on predicting short-term Global Horizontal Irradiance (GHI) using data from the Southern African Universities Radiometric Network (SAURAN) at the Univen Radiometric Station in South Africa. Various techniques were evaluated for their predictive accuracy, including Recurrent Neural Networks (RNN), Support Vector Regression (SVR), Gradient Boosting (GB), Random Forest (RF), Stacking Ensemble, and Double Nested Stacking (DNS). The results indicated that RNN performed the best in terms of Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) among the machine learning models. However, Stacking ensembles with XGBoost as the meta-model outperformed all individual models, improving accuracy by 67.06% in MAE and 22.28% in RMSE. DNS further enhanced accuracy, achieving a 93.05% reduction in MAE and an 88.54% reduction in RMSE compared to the best machine learning model, as well as a 78.89% decrease in MAE and an 85.27% decrease in RMSE compared to the best single stacking model. Furthermore, experimenting with the order of the DNS meta-model revealed that using RF as the first-level meta-model followed by XGBoost yielded the highest accuracy, showing a 47.39% decrease in MAE and a 61.35% decrease in RMSE compared to DNS with RF at both levels. These findings underscore the potential of advanced stacking techniques to significantly improve GHI forecasting. | |
| dc.description.sponsorship | DST-CSIR National e-Science Postgraduate Teaching and Training Platform (NEPTTP) | |
| dc.format.extent | 1 online resource (xi, 64 leaves): color illustrations | |
| dc.identifier.apacitation | Mugware, F. W. (2025). <i>Short-term forecasting of global horizontal irradiance using stacked ensemble machine learning alogorithms</i>. (). . Retrieved from | en_ZA |
| dc.identifier.chicagocitation | Mugware, Fhulufhelo Walter. <i>"Short-term forecasting of global horizontal irradiance using stacked ensemble machine learning alogorithms."</i> ., , 2025. | en_ZA |
| dc.identifier.citation | Mugware, F.W. 2025. Short-term forecasting of global horizontal irradiance using stacked ensemble machine learning alogorithms. . . | en_ZA |
| dc.identifier.ris | TY - Dissertation AU - Mugware, Fhulufhelo Walter AB - In today’s world, where sustainable energy is essential for the planet’s survival, accurate solar energy forecasting is crucial. This study focused on predicting short-term Global Horizontal Irradiance (GHI) using data from the Southern African Universities Radiometric Network (SAURAN) at the Univen Radiometric Station in South Africa. Various techniques were evaluated for their predictive accuracy, including Recurrent Neural Networks (RNN), Support Vector Regression (SVR), Gradient Boosting (GB), Random Forest (RF), Stacking Ensemble, and Double Nested Stacking (DNS). The results indicated that RNN performed the best in terms of Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) among the machine learning models. However, Stacking ensembles with XGBoost as the meta-model outperformed all individual models, improving accuracy by 67.06% in MAE and 22.28% in RMSE. DNS further enhanced accuracy, achieving a 93.05% reduction in MAE and an 88.54% reduction in RMSE compared to the best machine learning model, as well as a 78.89% decrease in MAE and an 85.27% decrease in RMSE compared to the best single stacking model. Furthermore, experimenting with the order of the DNS meta-model revealed that using RF as the first-level meta-model followed by XGBoost yielded the highest accuracy, showing a 47.39% decrease in MAE and a 61.35% decrease in RMSE compared to DNS with RF at both levels. These findings underscore the potential of advanced stacking techniques to significantly improve GHI forecasting. DA - 2025-09-05 DB - ResearchSpace DP - Univen KW - Double Nested Stacking KW - GB KW - GHI KW - Machine learning KW - RNN KW - RF KW - SAURAN KW - Solar energy KW - Stacking ensemble KW - SVR LK - https://univendspace.univen.ac.za PY - 2025 T1 - Short-term forecasting of global horizontal irradiance using stacked ensemble machine learning alogorithms TI - Short-term forecasting of global horizontal irradiance using stacked ensemble machine learning alogorithms UR - ER - | en_ZA |
| dc.identifier.uri | https://univendspace.univen.ac.za/handle/11602/3075 | |
| dc.identifier.vancouvercitation | Mugware FW. Short-term forecasting of global horizontal irradiance using stacked ensemble machine learning alogorithms. []. , 2025 [cited yyyy month dd]. Available from: | en_ZA |
| dc.language.iso | en | |
| dc.subject | Double Nested Stacking | |
| dc.subject | GB | |
| dc.subject | GHI | |
| dc.subject | Machine learning | |
| dc.subject | RNN | |
| dc.subject | RF | |
| dc.subject | SAURAN | |
| dc.subject | Solar energy | |
| dc.subject | Stacking ensemble | |
| dc.subject | SVR | |
| dc.title | Short-term forecasting of global horizontal irradiance using stacked ensemble machine learning alogorithms | |
| dc.type | Dissertation |