[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-122702-en":3,"doc-seo-122702-105":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},122702,962075006959,"Anda","https://ap-avatar.wpscdn.com/avatar/e0002397efbe92a78e?_k=1776741047341049297",8,"Research & Report","Efficient Surrogate Models for Materials Science Simulations - Machine Learning-based Prediction of Microstructure Properties","Determining, understanding, and predicting structure–property relations is central across chemistry, biology, physics, engineering, and materials science, where spatial microstructure details shape resulting properties in complex, non-trivial ways. Traditional forward simulations are costly, so this work develops six machine learning surrogate-model techniques using two materials-science datasets. Experiments cover a 2D Ising model for magnetic domain formation and Cahn–Hilliard-based dual-phase microstructure evolution, assessing accuracy, robustness, and performance drivers. Feature engineering with domain knowledge is evaluated, and data availability and quality guidelines are provided.","arXiv :2309 .00305v 1 [ cs .LG] 1 Sep 2023  \nEfficient Surrogate Models for Materials Science Simulations: Machine Learning-based Prediction of Microstructure Properties  \nBinh Duong Nguyena,c , Pavlo Potapenkoa,d , Aytekin Dermicia,e , Kishan Govinda,f, Stefan Sandfelda,b,g  \naInstitute for Advanced Simulations – Materials Data Science and Informatics (IAS-9), Forschungszentrum Jülich GmbH, 52425  \nJülich, Germany  \nbChair of Materials Data Science and Materials Informatics, Faculty 5 – Georesources and Materials Engineering, RWTH Aachen  \nUniversity, 52056 Aachen, Germany  \ncfirst author’s email address: bi.nguyen@fz-juelich.de  \nd second author’s email [address: p.potapenko@fz-juelich.de](address: p.potapenko@fz-juelich.de)  \n[e](e third author)[ third author](e third author)’[s email address: a.demirci@fz-juelich.de](s email address: a.demirci@fz-juelich.de)  \nffourth author’s email address: [k.govind@fz-juelich.de](k.govind@fz-juelich.de)  \ng corresponding author’s email address: [s.sandfeld@fz-juelich.de](s.sandfeld@fz-juelich.de)  \nAbstract  \nDetermining, understanding, and predicting the so-called structure-property relation is an important task in many scientific disciplines, such as chemistry, biology, meteorology, physics, engineering, and materials science. Structure refers to the spatial distribution of, e.g., substances, material, or matter in general, while property is a resulting characteristic that usually depends in a non-trivial way on spatial details of the structure. Traditionally, forward simulations models have been used for such tasks. Recently, several machine learning algorithms have been applied in these scientific fields to enhance and accelerate simulation models or as surrogate models. In this work, we develop and investigate the applications of six machine learning techniques based on two different datasets from the domain of materials science: data from a two-dimensional Ising model for predicting the formation of magnetic domains and data representing the evolution of dual-phase microstructures from the Cahn-Hilliard model. We analyze the accuracy and robustness of all models and elucidate the reasons for the differences in their performances. The impact of including domain knowledge through tailored features is studied, and general recommendations based on the availability and quality of training data are derived from this.  \nKeywords: structure-properties relation, forward model, feature engineering, power spectrum density, convolutional neural network, support vector regression, Ising model, Cahn-Hilliard model  \nPreprint submitted to Machine Learning with Applications September 4, 2023  \n1. Introduction  \nStudying the (micro)structure-properties relation is an important task for many different scientific fields and on many different length scales, e.g., for meteorology with up to kilometer-sized features, for materials science on the nanometer scale or for biological or chemical systems on various length scales (Kohn et al., 2018) . Mathematically, the task is to find the map from a (one-, two-, or threedimensional) spatial distribution of values to a single (scalar, vectorial, or tensorial) value. For example, geological measurements of the three-dimensional structural details of the earth crust are accompanied by displacement measurements which represents an average, i.e., an effective property, and can help to understand the general mechanism for shallow earthquakes (Tarasov, 2019) . In the field of weather forecasting, spatial details such as the structure of clouds or streamlines ofthe airflow determine properties such as cloud top temperature and particle effective radius (Rosenfeld et al., 2008) . Biological structures consist of molecules and cells, that are permanently evolving. Their subcellular interactions give rise to complex properties such as transport properties or how cells age (Li et al., 2021) . In the domain of material science and in particular, with regards to metallic","cbCaimfm3GNjahsA","https://ap.wps.com/l/cbCaimfm3GNjahsA","pdf",17129694,1,31,"English","en",105,"# Introduction\n## Structure–property relation and microstructure\n## Computational challenges in forward simulations\n## Machine learning surrogate models\n# Methods\n## Datasets: Ising and Cahn–Hilliard\n## Six machine learning techniques\n## Domain knowledge via tailored features\n# Results\n## Accuracy and robustness comparison\n## Reasons for performance differences\n## Impact of feature engineering\n# Recommendations\n## Training data availability and quality guidelines","[{\"question\":\"What problem does the work address in materials science simulations?\",\"answer\":\"It targets predicting the structure–property relation, mapping spatial microstructure information to resulting material properties that depend on detailed structure in a non-trivial way.\"},{\"question\":\"Which two datasets are used to evaluate the machine learning surrogate models?\",\"answer\":\"One dataset comes from a 2D Ising model for predicting magnetic domain formation, and the other represents dual-phase microstructure evolution generated from the Cahn–Hilliard model.\"},{\"question\":\"How does the study evaluate the benefit of domain knowledge?\",\"answer\":\"It investigates how including domain knowledge through tailored features affects accuracy and robustness, and then uses the findings to derive recommendations tied to training data availability and quality.\"}]","Efficient Surrogate Models for Materials Science Simulations - Machine Learning-based Prediction of Microstructure Properties | PDF",1785812373,78,{"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},"efficient-surrogate-models-for-materials-science-simulations-machine-learning-based-prediction-of-microstructure-properties","",{"@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/efficient-surrogate-models-for-materials-science-simulations-machine-learning-based-prediction-of-microstructure-properties/122702/",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 problem does the work address in materials science simulations?","Question",{"text":75,"@type":76},"It targets predicting the structure–property relation, mapping spatial microstructure information to resulting material properties that depend on detailed structure in a non-trivial way.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Which two datasets are used to evaluate the machine learning surrogate models?",{"text":80,"@type":76},"One dataset comes from a 2D Ising model for predicting magnetic domain formation, and the other represents dual-phase microstructure evolution generated from the Cahn–Hilliard model.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the study evaluate the benefit of domain knowledge?",{"text":84,"@type":76},"It investigates how including domain knowledge through tailored features affects accuracy and robustness, and then uses the findings to derive recommendations tied to training data availability and quality.","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"]