[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127035-en":3,"doc-seo-127035-105":30,"detail-sidebar-cat-0-en-105":92},{"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":20,"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},127035,962084928904,"Asher","https://ap-avatar.wpscdn.com/davatar_155a257f0dc6eb9ab79c44ca47cae57d",8,"Research & Report","Machine Learning for Fuel Cell Degradation - Report","Growing demand for sustainable energy drives the development of fuel cells as reliable energy-storage technologies. This thesis applies machine learning to predict and mitigate degradation in Solid Oxide Fuel Cells (SOFCs), building models that forecast voltage-drop and overpotential patterns to support maintenance planning and improve operational efficiency. Gaussian Process Regression models are developed, validated, and compared across datasets, including fluctuating-operation training, epsilon upgrades, and active learning updates. Model refinement and simulation-code adjustments enable clearer degradation observability, while results demonstrate improved predictive capability for extending lifespan and reducing operational costs.","Machine Learning for Fuel Cell Degradation  \nDocument:  \nReport  \nAuthor:  \nMarc Serna Jos´e  \nDirector:  \nXavi Salue˜na Berna  \nDegree:  \nM`aster en Enginyeria Industrial (MUEI)  \nConvocation:  \nDecember 2024  \nAbstract  \nThe increasing demand for sustainable energy solutions necessitates the advancement of technologies like fuel cells, which are integral to energy storage systems. This thesis explores the application of machine learning (ML) techniques to predict and mitigate degradation in fuel cells, specifically Solid Oxide Fuel Cells (SOFCs) . By leveraging ML, the research aims to develop predictive models that can accurately forecast degradation patterns, optimize maintenance schedules, and enhance the operational efficiency of SOFCs.  \nThe study involves the development and validation of Gaussian Process Regression (GPR) models, trained with various operational data to predict short-term voltage drops and overpotential in SOFCs. Different models are compared, including those trained with fluctuating operational data and upgraded with epsilon improvements to enhance prediction accuracy. The integration of active learning techniques further refines these models by iterative updating predictions based on new data, thus improving robustness and reliability. Model D was improved into Model D+, both predicting the difference between voltages in different time steps, but still needed a paradigm shift in order to be able to put it to use, as different operational conditions in time make it hard to be able to see through the voltage difference the degradation in fuel cells. Then, the SOFC simulation code was modified in order to get more data information between time steps and get enough data information to compare the SOFC at each time step in order to see degradation directly into voltage drop, so we applied this idea into Model E, even though we need more data to train this Model in order to validate it. In parallel with this we developed Model F keeping another idea in mind. Using the big amount of data already used in Model D, even though we had the operational conditions constant change, we targeted overpotential as the predicted feature, knowing that overpotential is one of the parameters we are interested in reducing in order to mitigate degradation in fuel cells. This was explored in Model G even with more detail as we divided the different overpotentials a SOFC experiments and predicted the ones that are most directly involved with degradation.  \nResults indicate that the ML models developed can effectively predict degradation, offering a proactive approach to maintenance that extends the lifespan of fuel cells and reduces operational costs. The findings underscore the potential of integrating ML into energy management systems, paving the way for more efficient, reliable, and sustainable energy solutions. This research contributes to the broader goal of achieving an AI-driven virtual power plant, enhancing the management of energy storage systems, and supporting the transition to a more sustainable energy future.  \nIndex  \n1 Introduction 5  \n1.1 Enegy Storage systems ........................................ 5  \n1.2 Inspiration .............................................. 6  \n1.3 Project goal and motivation ..................................... 6  \n1.4 ViPES2X ............................................... 7  \n2 Solid oxide fuel cells degradation 9  \n2.1 SOFC equations ........................................... 10  \n2.2 Degradation modelling ........................................ 11  \n2.2.1 Degradation in the anode .................................. 11  \n2.2.2 Degradation in the cathode ................................. 12  \n2.2.3 Degradation in the interconnect ............................... 12  \n2.3 Molar balance equations ....................................... 12  \n2.4 Species mass transport equations .................................. 14  \n2.5 Heat transfer equations .......................................","cbCaiiC0BZykfEfP","https://ap.wps.com/l/cbCaiiC0BZykfEfP","pdf",13078407,1,106,"English","en",105,"# Introduction\n## Energy storage systems\n## Inspiration\n## Project goal and motivation\n## ViPES2X\n# Solid oxide fuel cells degradation\n## SOFC equations\n## Degradation modelling\n## Molar balance equations\n## Species mass transport equations\n## Heat transfer equations\n# Prediction models with machine learning\n## GPR models\n## Radial basis function\n## Adam Optimizer\n## Active learning\n# Short-term prediction model for assisting real-time operation of the SOFC\n## Operating variables and parameters\n## Model D: Training GPR with fluctuating operation short-term data and voltage drop target value\n## Model D+: Training GPR with fluctuating operation short-term data Epsilon upgrade\n## Model E: Training GPR with fluctuating operation short-term data and real drop voltage of SOFC","[{\"question\":\"What is the main goal of this thesis?\",\"answer\":\"To apply machine learning techniques to predict and mitigate degradation in Solid Oxide Fuel Cells, enabling proactive maintenance and better operational efficiency.\"},{\"question\":\"Which machine learning approach is central to the prediction models?\",\"answer\":\"Gaussian Process Regression (GPR) models are developed, trained on operational data, and used to predict short-term voltage drops and overpotential.\"},{\"question\":\"How do the models improve prediction over earlier versions?\",\"answer\":\"Models are upgraded through epsilon improvements and refined iteratively with active learning, while simulation-code modifications generate richer between-time-step data to better observe degradation effects.\"}]","Machine Learning for Fuel Cell Degradation - Report | PDF",1785936472,267,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"machine-learning-for-fuel-cell-degradation-report","",{"@graph":36,"@context":86},[37,54,69],{"@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/machine-learning-for-fuel-cell-degradation-report/127035/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-21","2026-08-05",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What is the main goal of this thesis?","Question",{"text":76,"@type":77},"To apply machine learning techniques to predict and mitigate degradation in Solid Oxide Fuel Cells, enabling proactive maintenance and better operational efficiency.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"Which machine learning approach is central to the prediction models?",{"text":81,"@type":77},"Gaussian Process Regression (GPR) models are developed, trained on operational data, and used to predict short-term voltage drops and overpotential.",{"name":83,"@type":74,"acceptedAnswer":84},"How do the models improve prediction over earlier versions?",{"text":85,"@type":77},"Models are upgraded through epsilon improvements and refined iteratively with active learning, while simulation-code modifications generate richer between-time-step data to better observe degradation effects.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]