{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,15]],"date-time":"2026-01-15T11:30:02Z","timestamp":1768476602084,"version":"3.49.0"},"reference-count":79,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2025,8,7]],"date-time":"2025-08-07T00:00:00Z","timestamp":1754524800000},"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>Load losses negatively impact the reliability of power stations, leading to plant failures. To support the decision-making of improving plant reliability, we experimented with six machine learning classifiers to find the model combination that produces the best prediction performance, called the Explainable Multilayer Stack Ensemble. We applied a five-year dataset from six power stations. Since the dataset is highly imbalanced with the positive class dominant, class weights are calculated and assigned to reduce bias toward the majority class. The best parameters are determined through a randomized search with cross-validation and applied to train the models. The Explainable Multilayer Stack Ensemble performed better than the individual models, with a further improvement by excluding the Gaussian Na\u00efve Bayes in the second layer since it produced high false negatives. We demonstrate that when handling a highly imbalanced dataset, balanced accuracy, Receiver Operating Characteristics, and Precision-Recall Area Under the Curve provide a more reliable evaluation of model performance than focusing solely on standard evaluation metrics, such as accuracy, precision, and recall. Moreover, by excluding a poor-performing classifier from ensemble, we optimized the prediction process, and further enhanced overall performance.<\/jats:p>","DOI":"10.3389\/frai.2025.1592492","type":"journal-article","created":{"date-parts":[[2025,8,7]],"date-time":"2025-08-07T14:07:20Z","timestamp":1754575640000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Classification prediction of load losses in power stations using machine learning multilayer stack ensemble"],"prefix":"10.3389","volume":"8","author":[{"given":"Bathandekile M.","family":"Boshoma","sequence":"first","affiliation":[]},{"given":"Oluwole S.","family":"Akinola","sequence":"additional","affiliation":[]},{"given":"Peter","family":"Olukanmi","sequence":"additional","affiliation":[]}],"member":"1965","published-online":{"date-parts":[[2025,8,7]]},"reference":[{"key":"ref1","first-page":"1","author":"Abaas","year":"2022"},{"key":"ref2","doi-asserted-by":"publisher","first-page":"4528","DOI":"10.3390\/app14114528","article-title":"Digitalization processes in distribution grids: a comprehensive review of strategies and challenges","volume":"14","author":"Aghahadi","year":"2024","journal-title":"Appl. Sci."},{"key":"ref3","doi-asserted-by":"publisher","first-page":"875","DOI":"10.1080\/1206212X.2021.1974663","article-title":"Grid search in hyperparameter optimization of machine learning models for prediction of HIV\/AIDS test results","volume":"44","author":"Belete","year":"2022","journal-title":"Int. J. Comput. Appl."},{"key":"ref4","doi-asserted-by":"publisher","first-page":"13685","DOI":"10.3390\/su142013685","article-title":"Classification of cotton genotypes with mixed continuous and categorical variables: application of machine learning models","volume":"14","author":"Bishnoi","year":"2022","journal-title":"Sustainability"},{"key":"ref5","doi-asserted-by":"publisher","first-page":"1145","DOI":"10.1016\/S0031-3203(96)00142-2","article-title":"The use of the area under the ROC curve in the evaluation of machine learning algorithms","volume":"30","author":"Bradley","year":"1997","journal-title":"Pattern Recogn."},{"key":"ref6","author":"Brossa Dachs","year":"2018"},{"key":"ref7","doi-asserted-by":"publisher","first-page":"15262","DOI":"10.1609\/aaai.v35i17.17791","article-title":"Topological machine learning methods for power system responses to contingencies","volume":"35","author":"Bush","year":"2021","journal-title":"Proc. AAAI Conf. Artificial Intelligence"},{"key":"ref8","first-page":"518","author":"\u010ca\u010dkovi\u0107","year":"2016"},{"key":"ref9","volume-title":"Power failure Cascade prediction using machine learning","author":"Chadaga","year":"2023"},{"key":"ref10","doi-asserted-by":"publisher","first-page":"113761","DOI":"10.1016\/j.eswa.2020.113761","article-title":"A Q-learning agent for automated trading in equity stock markets","volume":"163","author":"Chakole","year":"2021","journal-title":"Expert Syst. Appl."},{"key":"ref11","doi-asserted-by":"publisher","first-page":"164","DOI":"10.1214\/aoms\/1177730978","article-title":"Relative accuracy of systematic and stratified random samples for a certain class of populations","volume":"17","author":"Cochran","year":"1946","journal-title":"Ann. Math. Stat."},{"key":"ref12","doi-asserted-by":"publisher","first-page":"38","DOI":"10.1609\/aimag.v42i2.15097","article-title":"Day-ahead forecasting of losses in the distribution network","volume":"42","author":"Dalal","year":"2021","journal-title":"AI Mag."},{"key":"ref13","first-page":"34","author":"Estabrooks","year":"2001"},{"key":"ref14","doi-asserted-by":"publisher","first-page":"137","DOI":"10.1007\/s11222-009-9153-8","article-title":"Estimation of prediction error by using K-fold cross-validation","volume":"21","author":"Fushiki","year":"2011","journal-title":"Stat. Comput."},{"key":"ref15","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s41044-016-0014-0","article-title":"Big data preprocessing: methods and prospects","volume":"1","author":"Garc\u00eda","year":"2016","journal-title":"Big Data Anal."},{"key":"ref16","doi-asserted-by":"publisher","first-page":"320","DOI":"10.1080\/01621459.1975.10479865","article-title":"The predictive sample reuse method with applications","volume":"70","author":"Geisser","year":"1975","journal-title":"J. Am. Stat. Assoc."},{"key":"ref17","doi-asserted-by":"publisher","first-page":"1894","DOI":"10.1002\/er.3957","article-title":"Future security of power supply in Germany\u2014the role of stochastic power plant outages and intermittent generation","volume":"42","author":"Gils","year":"2018","journal-title":"Int. J. Energy Res."},{"key":"ref18","first-page":"1","author":"Grando","year":"2019"},{"key":"ref19","doi-asserted-by":"publisher","first-page":"694","DOI":"10.1108\/IJRDM-09-2015-0140","article-title":"The digitalization of retailing: an exploratory framework","volume":"44","author":"Hagberg","year":"2016","journal-title":"Int. J. Retail Distrib. Manag."},{"key":"ref20","doi-asserted-by":"publisher","first-page":"1013","DOI":"10.13067\/JKIECS.2022.17.6.1013","article-title":"Regional analysis of load loss in power distribution lines based on Smartgrid big data","volume":"17","author":"Han-Min","year":"2022","journal-title":"J. Korea Institute Electron. Commun. Sci."},{"key":"ref21","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1186\/s40537-019-0274-4","article-title":"Severely imbalanced big data challenges: investigating data sampling approaches","volume":"6","author":"Hasanin","year":"2019","journal-title":"J. Big Data"},{"key":"ref22","doi-asserted-by":"crossref","DOI":"10.1002\/9781118646106","volume-title":"Imbalanced learning: Foundations, algorithms, and applications","author":"He","year":"2013"},{"key":"ref23","doi-asserted-by":"publisher","first-page":"108773","DOI":"10.1016\/j.ijepes.2022.108773","article-title":"Multi-stage hybrid energy management strategy for reducing energy abandonment and load losses among multiple microgrids","volume":"148","author":"Hou","year":"2023","journal-title":"Int. J. Electr. Power Energy Syst."},{"key":"ref24","volume-title":"The impact of digital transformation on the electric power industry: An explorative study of the largest norwegian distribution system operators","author":"Hustoft","year":"2019"},{"key":"ref25","doi-asserted-by":"publisher","first-page":"19","DOI":"10.17705\/1CAIS.00819","article-title":"Data mining; a conceptual overview","volume":"8","author":"Jackson","year":"2002","journal-title":"Commun. Assoc. Inf. Syst."},{"key":"ref26","volume-title":"Machine learning with scikit-learn quick start guide: Classification, regression, and clustering techniques in Python","author":"Jolly","year":"2018"},{"key":"ref27","doi-asserted-by":"publisher","first-page":"400","DOI":"10.3390\/su12010400","article-title":"Evaluation of nitrate load estimations using neural networks and canonical correlation analysis with k-fold cross-validation","volume":"12","author":"Jung","year":"2020","journal-title":"Sustainability"},{"key":"ref28","doi-asserted-by":"publisher","first-page":"4047","DOI":"10.24996\/ijs.2022.63.9.33","article-title":"Network traffic prediction based on boosting learning","author":"Kashmoola","year":"2022","journal-title":"Iraqi J. Sci."},{"key":"ref29","doi-asserted-by":"publisher","first-page":"21582440211047576","DOI":"10.1177\/21582440211047576","article-title":"Digital transformation: an overview of the current state of the art of research","volume":"11","author":"Kraus","year":"2021","journal-title":"SAGE Open"},{"key":"ref30","doi-asserted-by":"publisher","first-page":"1175","DOI":"10.1016\/j.matpr.2022.04.984","article-title":"Failure prediction of turbines using machine learning algorithms","volume":"66","author":"Kumar","year":"2022","journal-title":"Mater Today Proc"},{"key":"ref31","doi-asserted-by":"publisher","first-page":"2259","DOI":"10.1109\/TPWRS.2021.3119237","article-title":"A machine learning-based vulnerability analysis for cascading failures of integrated power-gas systems","volume":"37","author":"Li","year":"2021","journal-title":"IEEE Trans. Power Syst."},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1701.07274","article-title":"Deep reinforcement learning: An overview","author":"Li","year":"2017","journal-title":"arXiv preprint"},{"key":"ref33","doi-asserted-by":"publisher","first-page":"4698656","DOI":"10.1155\/2022\/4698656","article-title":"Stock trading strategies based on deep reinforcement learning","volume":"2022","author":"Li","year":"2022","journal-title":"Sci. Program."},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/EPEC48502.2020.9320125","article-title":"Real-time prediction of cascading failures in power systems","author":"Mahgoub","year":"2021","journal-title":"Univ. Saskatchewan"},{"key":"ref35","first-page":"1","author":"Mahgoub","year":"2020"},{"key":"ref36","first-page":"1","author":"Majzoobi","year":"2017"},{"key":"ref37","volume-title":"Machine learning, explained","author":"Management","year":"2021"},{"key":"ref38","doi-asserted-by":"publisher","first-page":"2009","DOI":"10.1007\/s00180-020-00999-9","article-title":"What is an optimal value of k in k-fold cross-validation in discrete Bayesian network analysis?","volume":"36","author":"Marcot","year":"2021","journal-title":"Comput. Stat."},{"key":"ref39","doi-asserted-by":"crossref","DOI":"10.1201\/9781420067194","volume-title":"Machine learning: An algorithmic perspective","author":"Marsland","year":"2011"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1051\/matecconf\/202134700011","article-title":"Development and validation of a robust integrated thermal power plant model for load loss analysis and identification","volume":"347","author":"Marx","year":"2021","journal-title":"MATEC Web Conf."},{"key":"ref41","doi-asserted-by":"publisher","first-page":"110","DOI":"10.3390\/data4030110","article-title":"Reinforcement learning in financial markets","volume":"4","author":"Meng","year":"2019","journal-title":"Data"},{"key":"ref42","volume-title":"Machine learning","author":"Mitchell","year":"1997"},{"key":"ref43","first-page":"243","author":"Mohammed","year":"2020"},{"key":"ref44","author":"Morakanyane","year":"2017"},{"key":"ref45","doi-asserted-by":"publisher","first-page":"72343","DOI":"10.1109\/ACCESS.2023.3294472","article-title":"Online identification of cascading events in power systems with renewable generation using measurement data and machine learning","volume":"11","author":"Nakas","year":"2023","journal-title":"IEEE Access"},{"key":"ref46","doi-asserted-by":"crossref","DOI":"10.1007\/978-3-030-77696-1","volume-title":"Application of machine learning and deep learning methods to power system problems","author":"Nazari-Heris","year":"2021"},{"key":"ref47","doi-asserted-by":"publisher","first-page":"2610","DOI":"10.31838\/jcr.07.19.322","article-title":"A review on machine learning (feature selection, classification and clustering) approaches of big data mining in different area of research","volume":"7","author":"Neeraj","year":"2020","journal-title":"J. Crit. Rev."},{"key":"ref48","author":"Nist","year":"2015"},{"key":"ref49","doi-asserted-by":"publisher","first-page":"61","DOI":"10.5815\/ijitcs.2021.06.05","article-title":"Performance of machine learning algorithms with different K values in K-fold crossvalidation","volume":"13","author":"Nti","year":"2021","journal-title":"Int. J. Inform. Technol. Computer Sci."},{"key":"ref50","doi-asserted-by":"publisher","first-page":"7251","DOI":"10.3390\/en16217251","article-title":"A stochastic load forecasting approach to prevent transformer failures and power quality issues amid the evolving electrical demands facing utilities","volume":"16","author":"O\u2019donnell","year":"2023","journal-title":"Energies"},{"key":"ref51","doi-asserted-by":"publisher","first-page":"155","DOI":"10.5897\/IJPS2014.4226","article-title":"Egbin power station generator availability and unit performance studies","volume":"10","author":"Ogieva","year":"2015","journal-title":"Int. J. Phys. Sci."},{"key":"ref52","doi-asserted-by":"publisher","first-page":"41281","DOI":"10.1109\/ACCESS.2023.3263490","article-title":"A comparative study between Bayesian network and hybrid statistical predictive models for proactive power system network resilience enhancement operational planning","volume":"11","author":"Omogoye","year":"2023","journal-title":"IEEE Access"},{"key":"ref53","first-page":"255","author":"Pavlyshenko","year":"2018"},{"key":"ref54","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1016\/j.eswa.2018.02.032","article-title":"Trading financial indices with reinforcement learning agents","volume":"103","author":"Pendharkar","year":"2018","journal-title":"Expert Syst. Appl."},{"key":"ref55","author":"Plevris","year":"2022"},{"key":"ref56","volume-title":"Booklet 3, fundamentals and systematics of availability for thermal power plants","author":"Powertech","year":"2008"},{"key":"ref57","first-page":"35","author":"Ray","year":"2019"},{"key":"ref58","first-page":"297","author":"Ruiz-Chavez","year":"2018"},{"key":"ref59","doi-asserted-by":"publisher","first-page":"1956","DOI":"10.3390\/app13031956","article-title":"An overview of machine learning, deep learning, and reinforcement learning-based techniques in quantitative finance: recent progress and challenges","volume":"13","author":"Sahu","year":"2023","journal-title":"Appl. Sci."},{"key":"ref60","doi-asserted-by":"publisher","first-page":"206","DOI":"10.1147\/rd.441.0206","article-title":"Some studies in machine learning using the game of checkers","volume":"44","author":"Samuel","year":"2000","journal-title":"IBM J. Res. Dev."},{"key":"ref61","doi-asserted-by":"publisher","first-page":"388","DOI":"10.15547\/tjs.2017.s.01.065","article-title":"Digital transformation of business","volume":"15","author":"Schwertner","year":"2017","journal-title":"Trakia Journal of Science"},{"key":"ref62","first-page":"1","author":"Sharifzadeh","year":"2021"},{"key":"ref63","first-page":"1","author":"Shuvo","year":"2020"},{"key":"ref64","first-page":"1","author":"Shuvro","year":"2019"},{"key":"ref65","first-page":"1034","author":"Tamilarasi","year":"2020"},{"key":"ref66","author":"Tareq","year":"2024"},{"key":"ref67","doi-asserted-by":"publisher","first-page":"193","DOI":"10.5281\/zenodo.11420895","article-title":"The impact of digitization towards technological revolution in the power industry","volume":"6","author":"Ugboke","year":"2024","journal-title":"J. Energy Technol. Environ."},{"key":"ref68","volume-title":"Utilisation of machine learning in power transformer asset Management","author":"Ukkelberg","year":"2018"},{"key":"ref69","first-page":"1","author":"Uwamahoro","year":"2023"},{"key":"ref70","doi-asserted-by":"publisher","first-page":"104504","DOI":"10.1016\/j.engappai.2021.104504","article-title":"Machine learning applications in power system fault diagnosis: research advancements and perspectives","volume":"106","author":"Vaish","year":"2021","journal-title":"Eng. Appl. Artif. Intell."},{"key":"ref71","first-page":"39","author":"Vanwinckelen","year":"2012"},{"key":"ref72","doi-asserted-by":"publisher","first-page":"129409","DOI":"10.1016\/j.energy.2023.129409","article-title":"A novel multi-layer stacking ensemble wind power prediction model under Tensorflow deep learning framework considering feature enhancement and data hierarchy processing","volume":"286","author":"Wang","year":"2024","journal-title":"Energy"},{"key":"ref73","doi-asserted-by":"publisher","first-page":"1281368","DOI":"10.3389\/fenrg.2023.1281368","article-title":"Online load-loss risk assessment based on stacking ensemble learning for power systems","volume":"11","author":"Wang","year":"2023","journal-title":"Front. Energy Res."},{"key":"ref74","doi-asserted-by":"publisher","first-page":"1","DOI":"10.25008\/ijadis.v2i1.1204","article-title":"Evaluation of the K-nearest neighbor model with K-fold cross validation on image classification","volume":"9","author":"Wayahdi","year":"2020","journal-title":"Info"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1051\/e3sconf\/202123602017","article-title":"Construction of digital operation and maintenance system for new energy power generation enterprises","volume":"236","author":"Wenyu","year":"2021","journal-title":"E3S Web Conf."},{"key":"ref76","doi-asserted-by":"publisher","first-page":"2150025","DOI":"10.1142\/S0218213021500251","article-title":"Topological machine learning for mixed numeric and categorical data","volume":"30","author":"Wu","year":"2021","journal-title":"Int. J. Artificial Intell. Tools"},{"key":"ref77","doi-asserted-by":"publisher","first-page":"214","DOI":"10.1016\/j.neucom.2021.09.072","article-title":"Stock movement prediction via gated recurrent unit network based on reinforcement learning with incorporated attention mechanisms","volume":"467","author":"Xu","year":"2022","journal-title":"Neurocomputing"},{"key":"ref78","doi-asserted-by":"crossref","DOI":"10.1016\/B978-0-323-95879-0.50234-4","volume-title":"Predictive maintenance in the digital era","author":"Yeardley","year":"2022"},{"key":"ref79","doi-asserted-by":"publisher","first-page":"14115","DOI":"10.1109\/ACCESS.2022.3197165","article-title":"Reinforcement learning for stock prediction and high-frequency trading with T+ 1 rules","volume":"11","author":"Zhang","year":"2022","journal-title":"IEEE Access"}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1592492\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,7]],"date-time":"2025-08-07T14:07:24Z","timestamp":1754575644000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2025.1592492\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,8,7]]},"references-count":79,"alternative-id":["10.3389\/frai.2025.1592492"],"URL":"https:\/\/doi.org\/10.3389\/frai.2025.1592492","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,8,7]]},"article-number":"1592492"}}