[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-124464-en":3,"doc-seo-124464-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},124464,1099514068365,"Aurelia","https://ap-avatar.wpscdn.com/avatar/10000253d8d9f28188e?_k=1776742907772140068",8,"Research & Report","SHORT-TERM FORECASTING OF GLOBAL HORIZONTAL IRRADIANCE USING STACKED ENSEMBLE MACHINE LEARNING ALGORITHMS","Accurate short-term solar forecasting supports reliable planning for sustainable energy systems. This study predicts Global Horizontal Irradiance (GHI) using SAURAN radiometric data from the Univen Radiometric Station in South Africa, assessing recurrent neural networks (RNN), support vector regression (SVR), gradient boosting (GB), random forests (RF), stacking ensembles, and double nested stacking (DNS). RNN achieves the best single-model error, measured by MAE and RMSE. Stacked ensembles with XGBoost as meta-model outperform all individual models, and DNS further reduces MAE and RMSE substantially. DNS meta-model ordering experiments show RF followed by XGBoost yields the highest accuracy, highlighting stacked ensemble design as a practical route to improved GHI forecasting.","SHORT-TERM FORECASTING OF GLOBAL HORIZONTAL IRRADIANCE USING STACKED ENSEMBLE MACHINE LEARNING ALGORITHMS  \nBy  \nFhulufhelo Walter Mugware 22020643  \nMini dissertation for the Master of Science Degree in E-Science  \nin the  \nDepartment of Mathematical and Computational Sciences, Faculty of Science, Engineering and Agriculture University of Venda, Thohoyandou, Limpopo South Africa  \nSupervisor: Dr T Ravele Co-Supervisor: Prof C Sigauke  \nFebruary 16, 2025  \nDeclaration  \nI, Fhulufhelo Walter Mugware, with student number 22020643, affirm that this research is entirely my own creation and has not been presented for any academic qualification at any other educational institution. It does not incorporate the written work of others unless duly acknowledged and cited as appropriate.  \nSigned (Student): .. ....................... ..................... Date: ..1.6./0.2. /2. ..02.5. ............  \ni  \nAbstract  \nIn 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 RFas 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.  \nKeywords: Double Nested Stacking, GB, GHI, Machine learning, RNN, RF, SAURAN, Solar energy, Stacking ensemble, SVR.  \nii  \nDedication  \nThis work is dedicated to my late father, Ntshengedzeni Jameson Mugware, and my entire family.  \niii  \nAcknowledgment  \nI would like to start by expressing my gratitude to God for His unwavering support throughout this journey. There were times when I felt drained, but His love and guidance gave me the strength to persevere.  \nI also want to thank my supervisor, Dr.T. Ravele, for the continuous support and invaluable guidance, and my co-supervisor, Prof. C. Sigauke, for constantly motivating me to learn and grow. This project would not have been possible without their dedication and expertise.  \nAlso want to thank my sister, who helped me begin my academic journey, and my mother, Ntsieni Mugware, who has been nothing but supportive. I also want to thank my brother, Anza Mugware, who always pushes me to work hard and wants nothing but the best for me. Lastly, I offer my heartfelt dedication to my sister, Mpho Mugware, and my little brother, Arendwaho Mugware.  \nI also want to give special thanks to my fellow students, Ndivhuwo and Asakundwi from the e-Science group of 2023, for their continuous assistance and collaboration throughout this project. Lastly I want to sincerely thank the University of Venda for welcoming me in to their academic community and the DST-CSIR National e-Science Postgraduate Teaching and Training Platform (NEPTTP) for their vital financial support.  \niv  \nTable of contents  ","cbCaivO8HrYIrCqM","https://ap.wps.com/l/cbCaivO8HrYIrCqM","pdf",1870158,1,75,"English","en",105,"# Table of contents\n## 1 Introduction\n## 1.1 Background\n## 1.2 Problem Statement\n## 1.3 Purpose of the study\n## 1.3.1 Aim\n## 1.3.2 Objectives\n## 1.3.3 Significance of the Study\n## 1.3.4 Contribution to knowledge\n## 1.4 Scope of the Study\n## 2 Literature review\n## 2.1 Introduction\n## 2.2 An overview of GHI forecasting using machine learning algorithms\n## 2.2.1 State-of-the-art in short-term forecasting of GHI using machine learning\n## 2.3 An overview of forecasting using ensemble stacking techniques\n## 2.3.1 Research gap(s)\n## 2.3.2 Contribution from this work\n## 2.4 Conclusions from literature\n## 3 Methodology\n## 3.1 Introduction\n## 3.2 Data source\n## 3.3 Models\n## 3.3.1 Recurrent neural networks\n## 3.3.2 Support Vector Regression\n## 3.3.3 Random Forests\n## 3.3.4 Gradient Boosting Model\n## 3.3.5 Stacking ensemble\n## 3.3.6 Double nested stacking\n## 3.4 Bayesian optimisation\n## 3.5 Variable selection\n## 3.6 Metrics for evaluating forecasts\n## 3.7 Conclusion\n## 4 Results and Discussions\n## 4.1 Introduction\n## 4.2 Dataset description\n## 4.3 Software and packages","[{\"question\":\"What data source is used for the short-term GHI forecasting study?\",\"answer\":\"The study uses data from the Southern African Universities Radiometric Network (SAURAN) at the Univen Radiometric Station in South Africa.\"},{\"question\":\"Which machine learning models are compared in the research?\",\"answer\":\"The research evaluates RNN, SVR, gradient boosting, random forests, stacking ensembles, and double nested stacking (DNS).\"},{\"question\":\"How do stacked ensembles and DNS improve forecasting accuracy compared with single models?\",\"answer\":\"Stacking ensembles with XGBoost as the meta-model outperform all individual models. DNS further improves results by substantially reducing MAE and RMSE, and the best DNS ordering uses RF as the first-level meta-model followed by XGBoost.\"}]","SHORT-TERM FORECASTING OF GLOBAL HORIZONTAL IRRADIANCE USING STACKED ENSEMBLE MACHINE LEARNING ALGORITHMS | PDF",1785822467,189,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"short-term-forecasting-of-global-horizontal-irradiance-using-stacked-ensemble-machine-learning-algorithms","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/short-term-forecasting-of-global-horizontal-irradiance-using-stacked-ensemble-machine-learning-algorithms/124464/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What data source is used for the short-term GHI forecasting study?","Question",{"text":75,"@type":76},"The study uses data from the Southern African Universities Radiometric Network (SAURAN) at the Univen Radiometric Station in South Africa.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Which machine learning models are compared in the research?",{"text":80,"@type":76},"The research evaluates RNN, SVR, gradient boosting, random forests, stacking ensembles, and double nested stacking (DNS).",{"name":82,"@type":73,"acceptedAnswer":83},"How do stacked ensembles and DNS improve forecasting accuracy compared with single models?",{"text":84,"@type":76},"Stacking ensembles with XGBoost as the meta-model outperform all individual models. DNS further improves results by substantially reducing MAE and RMSE, and the best DNS ordering uses RF as the first-level meta-model followed by XGBoost.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]