{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T05:05:12Z","timestamp":1789362312483,"version":"build-2803163510"},"reference-count":35,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T00:00:00Z","timestamp":1789344000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>\n                    Global solar radiation is a fundamental component of the Earth's energy balance and plays a critical role in integrating renewable energy into electrical systems. However, its pronounced variability, particularly in high-altitude regions, constrains energy stability and complicates planning processes. This study evaluated regression-based and machine-learning models for 1-day-ahead prediction of global horizontal irradiance (GHI) in Juliaca, Peru (&amp;gt;3,800 m.a.s.l.), using 11 years (2010\u20132020) of daily NASA POWER data (\n                    <jats:italic>n<\/jats:italic>\n                    = 4,000 observations after cleaning; 80\/20 chronological train\/test split: training from 2010 to 2018 and testing from 2018 to 2020). During data auditing, we identified that the same-day clearness index (\n                    <jats:italic>K<\/jats:italic>\n                    <jats:sub>\n                      <jats:italic>T<\/jats:italic>\n                    <\/jats:sub>\n                    ) is algebraically related to the same-day GHI through the extraterrestrial radiation term, which would otherwise inflate the apparent predictive accuracy. Therefore, we restricted same-day predictors to cyclical solar geometry terms and used lagged specific humidity, clearness index, and irradiance values from the previous 1\u20133 days as legitimate autoregressive information. Under this leakage-free design, Linear Regression achieved the best performance (\n                    <jats:italic>R<\/jats:italic>\n                    <jats:sup>2<\/jats:sup>\n                    = 0.390, MAE = 0.677 kWh m\n                    <jats:sup>\u22122<\/jats:sup>\n                    d\n                    <jats:sup>\u22121<\/jats:sup>\n                    , and RMSE = 0.896 kWh m\n                    <jats:sup>\u22122<\/jats:sup>\n                    d\n                    <jats:sup>\u22121<\/jats:sup>\n                    ), narrowly outperforming gradient boosting (\n                    <jats:italic>R<\/jats:italic>\n                    <jats:sup>2<\/jats:sup>\n                    = 0.384) and polynomial regression (\n                    <jats:italic>R<\/jats:italic>\n                    <jats:sup>2<\/jats:sup>\n                    = 0.381). Random Forest and shallow neural networks followed closely, with all six models exhibiting remarkably comparable performance. Feature-importance and SHAP analyses for the Random Forest model showed that previous-day irradiance and previous-day (lag-1) humidity were the dominant predictors. This pattern may be consistent with short-persistence cloud processes; however, cloud observations are required to verify this interpretation directly. The modest but realistic explanatory power obtained here, together with the diagnostic procedure used to achieve it, offers a template for detecting and avoiding under-reported leakage risk in solar radiation forecasting studies that use clearness-index-derived predictors.\n                  <\/jats:p>","DOI":"10.3389\/frai.2026.1870055","type":"journal-article","created":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T04:35:55Z","timestamp":1789360555000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Predicting global horizontal irradiance using regression-based machine learning models in high-altitude environments"],"prefix":"10.3389","volume":"9","author":[{"given":"Milton Edward","family":"Humpiri-Flores","sequence":"first","affiliation":[{"name":"Escuela Profesional de Ingenier\u00eda de Sistemas, Universidad Peruana Uni\u00f3n","place":["Juliaca, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"David","family":"Mamani-Pari","sequence":"additional","affiliation":[{"name":"Escuela Profesional de Ingenier\u00eda de Sistemas, Universidad Peruana Uni\u00f3n","place":["Juliaca, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Danny","family":"L\u00e9vano","sequence":"additional","affiliation":[{"name":"Escuela Profesional de Ingenier\u00eda de Sistemas, Universidad Peruana Uni\u00f3n","place":["Juliaca, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Roel Dante","family":"Gomez Apaza","sequence":"additional","affiliation":[{"name":"Escuela Profesional de Ingenier\u00eda de Sistemas, Universidad Peruana Uni\u00f3n","place":["Juliaca, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Esteban","family":"Tocto-Cano","sequence":"additional","affiliation":[{"name":"Escuela Profesional de Ingenier\u00eda de Sistemas, Universidad Peruana Uni\u00f3n","place":["Juliaca, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"\u00c1ngel Rosendo","family":"Condori-Coaquira","sequence":"additional","affiliation":[{"name":"Escuela Profesional de Ingenier\u00eda de Sistemas, Universidad Peruana Uni\u00f3n","place":["Juliaca, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Javier Linkolk","family":"L\u00f3pez-Gonzales","sequence":"additional","affiliation":[{"name":"Escuela de Posgrado, Universidad Peruana Uni\u00f3n","place":["Lima, Peru"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,9,14]]},"reference":[{"key":"B1","doi-asserted-by":"crossref","first-page":"240","DOI":"10.3390\/sym11020240","article-title":"Time series ARIMA model for prediction of daily and monthly average global solar radiation: the case study of Seoul, South Korea","volume":"11","author":"Alsharif","year":"2019","journal-title":"Symmetry"},{"key":"B2","unstructured":"Peru: Highest Solar Radiation on the Planet\n          \n          2025"},{"key":"B3","doi-asserted-by":"crossref","first-page":"9644","DOI":"10.1038\/s41598-022-13652-w","article-title":"Comprehensive assessment, review, and comparison of AI models for solar irradiance prediction based on different time\/estimation intervals","volume":"12","author":"Bamisile","year":"2022","journal-title":"Sci. Rep"},{"key":"B4","doi-asserted-by":"publisher","first-page":"172","DOI":"10.3390\/forecast5010009","article-title":"Global solar radiation forecasting based on hybrid models with combinations of meteorological parameters","volume":"5","author":"Belmahdi","year":"2023","journal-title":"Forecasting"},{"key":"B5","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1023\/A:1010933404324","article-title":"Random forests","volume":"45","author":"Breiman","year":"2001","journal-title":"Mach. Learn"},{"key":"B6","doi-asserted-by":"publisher","first-page":"1922","DOI":"10.3390\/en16135029","article-title":"Arima models in solar radiation forecasting: a systematic review","volume":"16","author":"Chodakowska","year":"2023","journal-title":"Energies"},{"key":"B7","doi-asserted-by":"publisher","first-page":"398","DOI":"10.3390\/atmos16040398","article-title":"Evaluation of solar radiation prediction models using AI: a performance comparison in the high-potential region of Konya, T\u00fcrkiye","volume":"16","author":"Demir","year":"2025","journal-title":"Atmosphere"},{"key":"B8","doi-asserted-by":"publisher","first-page":"1189","DOI":"10.1214\/aos\/1013203451","article-title":"Greedy function approximation: a gradient boosting machine","volume":"29","author":"Friedman","year":"2001","journal-title":"Ann. Stat"},{"key":"B9","doi-asserted-by":"crossref","first-page":"48","DOI":"10.1016\/j.neucom.2015.02.078","article-title":"Hybrid machine learning forecasting of solar radiation values","volume":"176","author":"Gala","year":"2016","journal-title":"Neurocomputing"},{"key":"B10","doi-asserted-by":"publisher","first-page":"173958","DOI":"10.1016\/j.scitotenv.2024.173958","article-title":"Harnessing the power of hybrid deep learning algorithm for the estimation of global horizontal irradiance","volume":"943","author":"Gupta","year":"2024","journal-title":"Sci. Tot. Environ"},{"key":"B11","first-page":"770","article-title":"\u201cDeep residual learning for image recognition,\u201d","volume-title":"Proceedings of the IEEE conference on computer vision and pattern recognition (CVPR)","author":"He","year":"2016"},{"key":"B12","doi-asserted-by":"publisher","first-page":"1527","DOI":"10.1162\/neco.2006.18.7.1527","article-title":"A fast learning algorithm for deep belief nets","volume":"18","author":"Hinton","year":"2006","journal-title":"Neural Comput"},{"key":"B13","doi-asserted-by":"publisher","first-page":"127","DOI":"10.1007\/s12667-024-00657-9","article-title":"A lightweight time series method for prediction of solar radiation","volume":"17","author":"Hissou","year":"2024","journal-title":"Energy Syst"},{"key":"B14","volume-title":"Renewable Energy in Power Systems","author":"Infield","year":"2020"},{"key":"B15","year":"2020","journal-title":"World Energy Outlook 2020"},{"key":"B16","doi-asserted-by":"publisher","first-page":"10761","DOI":"10.1038\/s41598-025-95118-3","article-title":"An optimized deep learning based hybrid model for prediction of daily average global solar irradiance using CNN-SLSTM architecture","volume":"15","author":"Mariappan","year":"2025","journal-title":"Sci. Rep"},{"key":"B17","volume-title":"Estad\u00edstica aplicada a la investigaci\u00f3n cient\u00edfica","author":"Menacho-Chiok","year":"2014"},{"key":"B18","doi-asserted-by":"publisher","first-page":"075010","DOI":"10.1063\/5.0211723","article-title":"Forecasting daily solar radiation: an evaluation and comparison of machine learning algorithms","volume":"14","author":"Nadeem","year":"2024","journal-title":"AIP Adv"},{"key":"B19","unstructured":"Nasa Power Data Access Viewer\n          \n          2025"},{"key":"B20","doi-asserted-by":"publisher","first-page":"101512","DOI":"10.1016\/j.esd.2024.101512","article-title":"Harnessing open data for hourly power generation forecasting in newly commissioned photovoltaic power plants","volume":"81","author":"Nastic","year":"2024","journal-title":"Energy Sustain. Dev"},{"key":"B21","doi-asserted-by":"publisher","first-page":"79","DOI":"10.1007\/978-981-97-5441-0_8","article-title":"Solar radiation prediction in the UTEQ based on machine learning models","volume":"1055","author":"Oviedo","year":"2023","journal-title":"Lect. Notes Netw. Syst"},{"key":"B22","volume-title":"Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change","author":"Pachauri","year":"2014"},{"key":"B23","doi-asserted-by":"crossref","first-page":"279","DOI":"10.1016\/j.renene.2020.04.042","article-title":"Solar radiation prediction using recurrent neural network and artificial neural network: a case study with comparisons","volume":"156","author":"Pang","year":"2020","journal-title":"Renew. Energy"},{"key":"B24","first-page":"499","article-title":"Effect of humidity on the efficiency of solar cell (photovoltaic)","volume":"2","author":"Panjwani","year":"2014","journal-title":"Int. J. Eng. Res. Gen. Sci"},{"key":"B25","unstructured":"How to Calculate the Annual Solar Energy Output of a Photovoltaic System\n          \n          2025"},{"key":"B26","first-page":"69","volume":"30","author":"Quispe Huam\u00e1n","year":"2022"},{"key":"B27","doi-asserted-by":"publisher","first-page":"533","DOI":"10.1038\/323533a0","article-title":"Learning representations by back-propagating errors","volume":"323","author":"Rumelhart","year":"1986","journal-title":"Nature"},{"key":"B28","doi-asserted-by":"crossref","first-page":"3179","DOI":"10.3390\/en19133179","article-title":"Project risk assessment of renewable energy projects in electricity market structures: a systematic literature review","volume":"19","author":"Tampubolon","year":"2026","journal-title":"Energies"},{"key":"B29","doi-asserted-by":"publisher","first-page":"1097","DOI":"10.1016\/j.rser.2015.05.049","article-title":"Evaluation and performance comparison of different models for the estimation of solar radiation","volume":"50","author":"Teke","year":"2015","journal-title":"Renew. Sustain. Energy Rev"},{"key":"B30","doi-asserted-by":"crossref","first-page":"5436","DOI":"10.3390\/en16145436","article-title":"A review of state-of-the-art and short-term forecasting models for solar PV power generation","volume":"16","author":"Tsai","year":"2023","journal-title":"Energies"},{"key":"B31","volume-title":"Fundamentals of Solar Radiation","author":"Wald","year":"2018"},{"key":"B32","doi-asserted-by":"publisher","first-page":"1660","DOI":"10.3390\/en10101660","article-title":"Predictions of surface solar radiation on tilted solar panels using machine learning models: a case study of Tainan City, Taiwan","volume":"10","author":"Wei","year":"2017","journal-title":"Energies"},{"key":"B33","doi-asserted-by":"publisher","first-page":"131549","DOI":"10.1016\/j.energy.2024.131549","article-title":"Medium-term forecasting of global horizontal solar radiation in Brazil using machine learning-based methods","volume":"300","author":"Weyll","year":"2024","journal-title":"Energy"},{"key":"B34","doi-asserted-by":"publisher","first-page":"3215","DOI":"10.3390\/en16073215","article-title":"Simplified method for predicting hourly global solar radiation using extraterrestrial radiation and limited weather forecast parameters","volume":"16","author":"Yang","year":"2023","journal-title":"Energies"},{"key":"B35","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1016\/j.enbenv.2023.08.001","article-title":"Hourly solar radiation prediction using extreme gradient boosting with fog-haze effects","volume":"6","author":"Zhang","year":"2025","journal-title":"Energy Built Environ"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1870055\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T04:36:08Z","timestamp":1789360568000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1870055\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,14]]},"references-count":35,"alternative-id":["10.3389\/frai.2026.1870055"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1870055","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,14]]},"article-number":"1870055"}}