[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-127293-en":3,"doc-seo-127293-105":31,"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":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},127293,2336475104957,"Seraphina","https://ap-avatar.wpscdn.com/avatar/22000c4c6bd8a5076e1?x-image-process=image/resize,m_fixed,w_180,h_180&k=1786593998035447633",8,"Research & Report","A neural network machine-learning approach for characterising hydrogen trapping parameters from TDS experiments","Hydrogen trapping in metallic alloys is commonly characterised using Thermal Desorption Spectroscopy (TDS), yet converting TDS spectra into key parameters such as trap binding energies and densities remains difficult because the measurement is indirect. A multi-neural-network scheme is presented to identify trapping parameters directly from experimental spectra. The workflow trains two fully connected feed-forward networks with backpropagation on synthetic data only: a classification model for the number of trap types and a regression model for trap densities and binding energies, with architectures and preprocessing optimised to reduce training data. Results show strong predictive performance on tempered martensitic steels of different compositions.","International Journal of Hydrogen Energy 167 (2025) 150874  \n| A neural network machine-learning approach for characterising hydrogen trapping parameters from TDS experiments\u003Cbr>Nicoletta Marrani, Tim Hageman ∗, Emilio Martínez-Pañeda ∗\u003Cbr>Department of Engineering Science, University of Oxford, Oxford OX1 3PJ, UK |  |\n| --- | --- |\n| A R T I C L E I N F O | A B S T R A C T\u003Cbr>The hydrogen trapping behaviour of metallic alloys is generally characterised using Thermal Desorption Spectroscopy (TDS). However, as an indirect method, extracting key parameters (trap binding energies and densities) remains a significant challenge. To address these limitations, this work introduces a machine learning-based scheme for parameter identification from TDS spectra. A multi-Neural Network (NN) model is developed and trained exclusively on synthetic data to predict trapping parameters directly from experimental data. The model comprises two multi-layer, fully connected, feed-forward NNs trained with backpropagation. The first network (classification model) predicts the number of distinct trap types. The second network (regression model) then predicts the corresponding trap densities and binding energies. The NN architectures, hyperparameters, and data pre-processing were optimised to minimise the amount of training data. The proposed model demonstrated strong predictive capabilities when applied to three tempered martensitic steels of different compositions. The code developed is freely provided. |\n| Dataset link: [https://github.com/nicolettamarr](https://github.com/nicolettamarr)ani/TDS_ML_Approach.git, [https://mechmat.w](https://mechmat.w)[eb.ox.ac.uk/codes](eb.ox.ac.uk/codes) |  |\n| Keywords:\u003Cbr>Hydrogen\u003Cbr>Thermal desorption spectroscopy\u003Cbr>Trapping\u003Cbr>Parameter identification\u003Cbr>Machine learning\u003Cbr>Neural network |  |\n\n1. Introduction  \nTo transition away from fossil fuels as the primary energy source, the global energy industry has increasingly focused on developing low-carbon technologies. Alongside the widespread adoption of renewable energy, hydrogen is gaining prominence as both a fuel and an energy carrier [1–3]. Its natural abundance and low projected environmental impact position hydrogen as a promising solution fordecarbonising traditionally hard-to-abate industries [4,5]. However, the development of a hydrogen-based economy is hindered by hydrogen’s tendency to degrade the mechanical properties of structural materials —a phenomenon known as hydrogen embrittlement [6]. A detailed understanding of hydrogen embrittlement is necessary before existing energy infrastructure, such as gas pipelines, is adapted to transport hydrogen [7,8], with one of the main limitations being the characterisation of metal properties.  \nDespite extensive experimental and computational efforts to understand the mechanisms of hydrogen embrittlement, characterising the properties of metals and predicting their degradation due to hydrogen remains a challenge [9]. These predictions require a thorough examination of hydrogen–metal interactions, including ingress [10–13], diffusion through the lattice [14,15], and trapping at microstructural imperfections [16–18]. In particular, the rapid diffusion of hydrogen  \nthrough the metal lattice and its subsequent trapping at crystal defects have been identified as key factors influencing susceptibility to hydrogen embrittlement [6,19].  \nOne well-established technique for characterising hydrogen trapping behaviour in metallic alloys is Thermal Desorption Spectroscopy (TDS) [20], also referred to as Thermal Desorption Analysis (TDA). TDS analysis has been used extensively to understand how microstructural features—such as vacancies [21], dislocations [22–24], grain boundaries [25], voids [26–28], and precipitates [29–31]—interact with and retain hydrogen. Each of these features, commonly referred to as traps, is characterised by a unique binding energy and trap density, which TDS can quantify. The technique is valu","cbCaisB7ctJC1wtO","https://ap.wps.com/l/cbCaisB7ctJC1wtO","pdf",3268381,2,1,19,"English","en",105,"# Introduction\n## Hydrogen embrittlement and the role of hydrogen trapping\n## Thermal Desorption Spectroscopy (TDS) and trap parameters\n## Limitations of indirect parameter extraction and existing analysis approaches","[{\"question\":\"Why is extracting hydrogen trapping parameters from TDS experiments challenging?\",\"answer\":\"TDS is an indirect technique, so the desorption spectrum reflects the combined contribution of multiple traps, making parameter extraction and interpretation nontrivial.\"},{\"question\":\"How does the proposed multi-neural-network model work?\",\"answer\":\"It uses two feed-forward neural networks trained on synthetic data: a classification network predicts the number of distinct trap types, and a regression network predicts the corresponding trap densities and binding energies from TDS spectra.\"},{\"question\":\"What data were used to train the neural networks?\",\"answer\":\"The networks were trained exclusively on synthetic data, and model design and preprocessing were optimised to minimise the amount of required training data.\"}]","A neural network machine-learning approach for characterising hydrogen trapping parameters from TDS experiments | PDF",1785938151,48,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":29},"a-neural-network-machine-learning-approach-for-characterising-hydrogen-trapping-parameters-from-tds-experiments","",{"@graph":37,"@context":86},[38,54,69],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,48,51],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":20},"https://docshare.wps.com/document/","Document",{"item":49,"name":12,"@type":44,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":44,"position":53},"https://docshare.wps.com/document/a-neural-network-machine-learning-approach-for-characterising-hydrogen-trapping-parameters-from-tds-experiments/127293/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":42,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-22","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},"Why is extracting hydrogen trapping parameters from TDS experiments challenging?","Question",{"text":76,"@type":77},"TDS is an indirect technique, so the desorption spectrum reflects the combined contribution of multiple traps, making parameter extraction and interpretation nontrivial.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How does the proposed multi-neural-network model work?",{"text":81,"@type":77},"It uses two feed-forward neural networks trained on synthetic data: a classification network predicts the number of distinct trap types, and a regression network predicts the corresponding trap densities and binding energies from TDS spectra.",{"name":83,"@type":74,"acceptedAnswer":84},"What data were used to train the neural networks?",{"text":85,"@type":77},"The networks were trained exclusively on synthetic data, and model design and preprocessing were optimised to minimise the amount of required training data.","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":25},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":47,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":47,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":47,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":47,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":47,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":47,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":22,"doc_module":4,"doc_module_name":47,"category_name":137,"show_sort_weight":107,"slug":138},"General","general"]