[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-123390-en":3,"doc-seo-123390-105":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":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},123390,962075114101,"Seraphina","https://ap-avatar.wpscdn.com/avatar/e000253a75eb197efd?x-image-process=image/resize,m_fixed,w_180,h_180&k=1780044092746381165",8,"Research & Report","Accelerating Hasegawa–Wakatani simulations with machine learning for out-of-distribution predictions","Plasma turbulence simulation faces major computational bottlenecks due to coupled multi-scale dynamics. This study applies convolutional neural networks to accelerate Hasegawa–Wakatani turbulence simulations by learning closure terms for large eddy simulations. Models are trained on selected adiabatic coefficients and tested on unseen coefficients beyond the training range, using ground-truth data for multiple C values. Results show accurate particle-flux prediction and successful handling of initialization through state transfer, enabling fast convergence without costly direct simulations. ","Plasma Phys. Control. Fusion 67 (2025) 045018 (15pp) [https://doi.org/10.1088/1361-6587/adbb1c](https://doi.org/10.1088/1361-6587/adbb1c)  \nAccelerating Hasegawa–Wakatani simulations with machine learning for out-of-distribution predictions  \nV Artigues 1, ∗ 􀁂, Robin Greif2 􀁂 and F Jenko 1 􀁂  \n1 Max Planck Institute for Plasma Physics, Boltzmannstr. 2, 85748 Garching, Germany  \n2 University of Oxford, Beecroft Building, Parks Rd, Oxford OX1 3PU, United Kingdom  \nE-mail: [victor.artigues@ipp.mpg.de](victor.artigues@ipp.mpg.de)  \nReceived 10 December 2024, revised 19 February 2025 Accepted for publication 27 February 2025  \nPublished 14 March 2025  \nAbstract  \nSimulating plasma turbulence presents significant computational challenges due to the complex interplay of multi-scale dynamics. In this work, we investigate the use of convolutional neural networks to improve the efficiency of plasma turbulence simulations, focusing on the Hasegawa–Wakatani model. The networks are trained to learn the closure terms in large eddy simulations, providing a computationally cheaper alternative to the high-resolution numerical solvers for capturing the effects of high-frequency components. This study is the first to successfully apply machine learning to predict plasma behavior for adiabatic coefficients beyond the training range for the Hasegawa–Wakatani equations. We generate ground truth simulations for three values of the adiabatic coefficient (C ∈ {0 .2 , 1.0 , 5.0}), and train our models on one, or two. The evaluation is then performed on the remaining values. The models generalize well and accurately predict the particle flux up to a factor 5 outside the training range. Finally, we address a key challenge in machine-learning-accelerated plasma simulations—initialization—by starting simulations for previously unseen adiabatic coefficients C with states from existing simulationsat other known C values. This approach removes the need for expensive direct numerical simulations for initialization while maintaining physical accuracy and a fast convergence rate. Overall, the results highlight the model’s strong generalization capabilities and its potential for accelerating plasma turbulence simulations with more complex sets of parameters.  \nKeywords: plasma turbulence, Hasegawa–Wakatani, machine learning, convolutional neural networks, large eddy simulations  \n1. Introduction  \nSimulating plasma turbulence, especially in the context of magnetic confinement fusion research, presents significant challenges and opportunities. Achieving fast and accurate  \n∗  \nAuthor to whom any correspondence should be addressed.  \nOriginal Content from this work may be used under the  \nterms of the Creative Commons Attribution 4.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.  \nplasma turbulence simulations could revolutionize the preparation, interpretation, and optimization of fusion experiments well beyond current capabilities.  \nVery complex codes have been developed over the last decades [1–8] to capture the often delicate nonlinear dynamics underlying plasma turbulence, involving a wide range of scales in position-velocity space and time. Despite remarkable advances in both numerical algorithms and hardware performance, these simulations often require very large computational resources. As these codes evolve to bridge the gap between qualitative and quantitative predictions, reducing their cost becomes increasingly relevant. By lowering the cost, simulations that were once out of reach such as those involving larger  \n1 © 2025 The Author(s) . Published by IOP Publishing Ltd  \nfusion devices, better resolution of current cases, or the inclusion of more species would become feasible. Additionally, simulations that are currently possible but computationally expensive could be performed routinely, enabling parameterscans and optimization studies.  \nThere are severa","cbCaibTLcOzkqBjy","https://ap.wps.com/l/cbCaibTLcOzkqBjy","pdf",1441618,1,15,"English","en",105,"# Introduction\n## Plasma turbulence and computational challenges\n## Speed-up strategies and machine-learning acceleration\n## Focus on Hasegawa–Wakatani model and LES closure\n## Overview of the paper structure","[{\"question\":\"How does the approach accelerate Hasegawa–Wakatani plasma turbulence simulations?\",\"answer\":\"Convolutional neural networks learn closure terms for large eddy simulations, providing a cheaper alternative to high-resolution numerical solvers while retaining the impact of high-frequency components.\"},{\"question\":\"What does “out-of-distribution” mean in this study?\",\"answer\":\"The models are evaluated on adiabatic coefficients outside the range used for training, testing whether learned closures generalize to previously unseen parameter values.\"},{\"question\":\"How is the initialization challenge addressed for unseen adiabatic coefficients?\",\"answer\":\"Simulations for new coefficients start from states taken from existing simulations at other known coefficients, avoiding expensive direct numerical simulations for initialization while maintaining physical accuracy and fast convergence.\"}]","Accelerating Hasegawa–Wakatani simulations with machine learning for out-of-distribution predictions | PDF",1785816250,38,{"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},"accelerating-hasegawawakatani-simulations-with-machine-learning-for-out-of-distribution-predictions","",{"@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/accelerating-hasegawawakatani-simulations-with-machine-learning-for-out-of-distribution-predictions/123390/",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":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"How does the approach accelerate Hasegawa–Wakatani plasma turbulence simulations?","Question",{"text":75,"@type":76},"Convolutional neural networks learn closure terms for large eddy simulations, providing a cheaper alternative to high-resolution numerical solvers while retaining the impact of high-frequency components.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What does “out-of-distribution” mean in this study?",{"text":80,"@type":76},"The models are evaluated on adiabatic coefficients outside the range used for training, testing whether learned closures generalize to previously unseen parameter values.",{"name":82,"@type":73,"acceptedAnswer":83},"How is the initialization challenge addressed for unseen adiabatic coefficients?",{"text":84,"@type":76},"Simulations for new coefficients start from states taken from existing simulations at other known coefficients, avoiding expensive direct numerical simulations for initialization while maintaining physical accuracy and fast convergence.","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"]