[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119845-en":3,"doc-seo-119845-105":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},119845,8796095360427,"Lucas Martin","https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d",8,"Research & Report","Incremental machine learning-based accelerator for computational fluid dynamics simulations - thesis","The simulation of physicochemical processes using computational methods underpins engineering design across industries, yet persists with major accuracy limits and high computational complexity. These multiphysics systems are governed by nonlinear transport equations spanning disparate spatiotemporal scales within complex geometries, making many cases computationally expensive or infeasible. Machine learning offers promising acceleration through neural networks trained on multiphysics experiment and simulation data. This work reduces the gap between ML and practical engineering workflows using (1) a unified training-and-simulation framework and (2) incremental, parallel determination of neural network parameters to cut data collection and training time while increasing multiphysics simulation speedup.","Incremental machine learning-based accelerator for computational fluid dynamics simulations  \nby  \nSajeda Mokbel  \nA thesis  \npresented to the University of Waterloo  \nin fulfillment of the  \nthesis requirement for the degree of  \nMaster of Applied Science  \nin  \nChemical Engineering  \nWaterloo, Ontario, Canada, 2023  \n© Sajeda Mokbel 2023  \nAuthor’s Declaration  \nI hereby declare that I am the sole author of this thesis. This is a true copy of the thesis, including any required final revisions, as accepted by my examiners.  \nI understand that my thesis may be made electronically available to the public.  \nAbstract  \nThe simulation of physicochemical processes with computational methods is key for engineering design, with applications in a variety of industries, ranging from pharmaceuticals to aerodynamics. Despite its importance and widespread use, significant challenges related to the accuracy and computational complexity of these simulations remain prominent.  \nThese systems are governed by non-linear transport equations with physical and chemical processes occurring at different spatiotemporal scales in complex geometries. This leads to problems which are computationally expensive and often infeasible to solve. As such, reducing the computational complexity of multiphysics problems without compromising on accuracy is a central goal in the engineering community.  \nRecently, machine learning has proven to be a promising direction towards this goal. The availability of data from both multiphysics experiments and simulations have led to high-performing neural networks capable of accelerating traditional methods for solving multiphysics problems. Despite these hopeful results, there still exists a gap between machine learning and its optimal application in a realistic engineering design process. This work aims to bridge that gap through two main approaches.  \nThe first approach is by developing a framework which hosts neural network training and existing computational multiphysics software in a unified framework. The second approach is to incrementally determine optimal neural network parameters by running computational multiphysics problems and neural network training in parallel. This has shown to reduce data collection and training time while increasing the speedup of multiphysics simulations over increments.  \nAcknowledgements  \nI would first like to thank my supervisor, Dr. Nasser Mohieddin Abukhdeir, for his guidance, mentorship, and supervision throughout the entirety of this work. I would also like to thank Dr. Hector Budman for his guidance and advice throughout this project.  \nA special thanks to the wonderful COMPHYS group-Alex, Matthew, Arshia, James, Ittisak and Nicholas, thank you for your help, moral support, and friendship during this learning journey.  \nThank you to my sisters, Aseel and Abir, and to my dad, for their constant support.  \nDedication  \nTo my mom. Thank you for all your love and support.  \nTable of Contents  \nList of Figures viii  \nList of Tables xi  \n1 Introduction 1  \n1.1 Research Motivation .............................. 1  \n1.2 Objectives .................................... 3  \n1.3 Thesis Structure ................................. 4  \n2 Background 5  \n2.1 Computational Multiphysics in Chemical Engineering ............ 5  \n2.2 Navier-Stokes equations ............................ 6  \n2.3 Numerical methods ............................... 13  \n2.3.1 Finite Volume Method ......................... 13  \n2.4 Machine Learning ................................ 17  \n2.4.1 Feedforward neural network ...................... 18  \n2.4.2 Convolutional Neural Networks .................... 21  \n2.4.3 Graph Neural Networks ........................ 24  \n3 Literature Review 26  \n3.1 ML to infer the solution of a model ...................... 27  \n3.1.1 Convolutional neural networks ..................... 27  \n3.1.2 Graph neural networks ......................... 28  \n3.1.3 Physics-informed neural networks ..........","cbCail9MRA7tumy0","https://ap.wps.com/l/cbCail9MRA7tumy0","pdf",15503409,1,82,"English","en",105,"# 1 Introduction\n## 1.1 Research Motivation\n## 1.2 Objectives\n## 1.3 Thesis Structure\n# 2 Background\n## 2.1 Computational Multiphysics in Chemical Engineering\n## 2.2 Navier-Stokes equations\n## 2.3 Numerical methods\n## 2.4 Machine Learning\n# 3 Literature Review\n## 3.1 ML to infer the solution of a model\n## 3.2 ML for the development of hybrid solvers\n# 4 Research Methods\n## 4.1 Simulation of Incompressible Navier Stokes\n## 4.2 Neural Network Creation\n## 4.3 Data generation\n# 5 Results\n## 5.1 Validation of CFDNet\n## 5.2 Recursion\n## 5.3 Hyperparameter Optimization\n## 5.4 Implementation of an end-to-end framework\n# 6 Conclusions & Future Work","[{\"question\":\"What problem does this thesis address in computational fluid dynamics (CFD) simulations?\",\"answer\":\"It targets the persistent challenges of accuracy and high computational complexity in multiphysics CFD simulations, which can become computationally expensive or infeasible.\"},{\"question\":\"How does the proposed approach bridge the gap between machine learning and engineering design workflows?\",\"answer\":\"It uses two methods: a unified framework combining neural network training with existing computational multiphysics software, and an incremental strategy that runs simulation and neural network training in parallel to determine optimal parameters.\"},{\"question\":\"What benefits are reported for the incremental training strategy?\",\"answer\":\"The incremental parallel approach reduces data collection and training time and increases the speedup of multiphysics simulations across increments.\"}]","Incremental machine learning-based accelerator for computational fluid dynamics simulations - thesis | PDF",1785726619,207,{"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},"incremental-machine-learning-based-accelerator-for-computational-fluid-dynamics-simulations-thesis","",{"@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/incremental-machine-learning-based-accelerator-for-computational-fluid-dynamics-simulations-thesis/119845/",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","2026-08-03",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 this thesis address in computational fluid dynamics (CFD) simulations?","Question",{"text":75,"@type":76},"It targets the persistent challenges of accuracy and high computational complexity in multiphysics CFD simulations, which can become computationally expensive or infeasible.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the proposed approach bridge the gap between machine learning and engineering design workflows?",{"text":80,"@type":76},"It uses two methods: a unified framework combining neural network training with existing computational multiphysics software, and an incremental strategy that runs simulation and neural network training in parallel to determine optimal parameters.",{"name":82,"@type":73,"acceptedAnswer":83},"What benefits are reported for the incremental training strategy?",{"text":84,"@type":76},"The incremental parallel approach reduces data collection and training time and increases the speedup of multiphysics simulations across increments.","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"]