{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,31]],"date-time":"2026-01-31T01:42:50Z","timestamp":1769823770242,"version":"3.49.0"},"reference-count":45,"publisher":"IOP Publishing","issue":"2","license":[{"start":{"date-parts":[[2025,6,12]],"date-time":"2025-06-12T00:00:00Z","timestamp":1749686400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2025,6,12]],"date-time":"2025-06-12T00:00:00Z","timestamp":1749686400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Mach. Learn.: Sci. Technol."],"published-print":{"date-parts":[[2025,6,30]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>The direct simulation Monte Carlo (DSMC) is a widely used approach for studying aerodynamics effects of rarefied flows, but it is highly time-consuming and may exhibit statistical fluctuations. In this study, we propose an efficient aerodynamic prediction method based on convolutional neural networks (CNNs) to further explore the application of deep learning in improving the efficiency of DSMC for calculating the aerodynamics of rarefied flows. The method includes centroid aerodynamics forces prediction (CFP) and surface aerodynamic forces distribution prediction (SFP), both of which are trained using a dataset of free molecular flow around obstacles derived from DSMC simulations. The SFP is designed to bridge the gap between flow field and surface forces, with two characteristics extraction methods developed specifically for this purpose. Additionally, two data preprocessing methods are designed to suppress the statistical noise inherent in DSMC simulations. Both CFP and SFP have demonstrated optimal performance in terms of accuracy and resistance to overfitting, achieving considerable predictive accuracy. The SFP exhibits a significant speedup, enabling real-time prediction of aerodynamic distributions from flow field. The results demonstrate that the proposed CNN-based approach offers a promising solution for the efficient calculation of aerodynamic forces in rarefied flows, and provide a robust foundation for ongoing development.<\/jats:p>","DOI":"10.1088\/2632-2153\/addf10","type":"journal-article","created":{"date-parts":[[2025,5,30]],"date-time":"2025-05-30T22:54:01Z","timestamp":1748645641000},"page":"025060","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["Efficient prediction of aerodynamic forces in rarefied flow using convolutional neural network based multi-process method"],"prefix":"10.1088","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0009-0005-9120-7959","authenticated-orcid":true,"given":"Haifeng","family":"Huang","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Guobiao","family":"Cai","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Chuanfeng","family":"Wei","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Baiyi","family":"Zhang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiang","family":"Cui","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yongjia","family":"Zhao","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Huiyan","family":"Weng","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6022-1441","authenticated-orcid":true,"given":"Weizong","family":"Wang","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3752-038X","authenticated-orcid":true,"given":"Lihui","family":"Liu","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6346-7981","authenticated-orcid":false,"given":"Bijiao","family":"He","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"266","published-online":{"date-parts":[[2025,6,12]]},"reference":[{"key":"mlstaddf10bib1","doi-asserted-by":"publisher","first-page":"706","DOI":"10.3390\/aerospace9110706","article-title":"A review of research on the vacuum plume","volume":"9","author":"Cai","year":"2022","journal-title":"Aerospace"},{"key":"mlstaddf10bib2","doi-asserted-by":"publisher","first-page":"1058","DOI":"10.1007\/s11431-016-9015-7","article-title":"Investigation on aerodynamic force effect of vacuum plumes using pressure-sensitive paint technique and CFD-DSMC solution","volume":"60","author":"Wu","year":"2017","journal-title":"Sci. China Technol. Sci."},{"key":"mlstaddf10bib3","doi-asserted-by":"publisher","first-page":"1536","DOI":"10.2514\/1.A35362","article-title":"Impact of stagnation temperature and nozzle configuration on rarefied jet plume interactions","volume":"59","author":"Agir","year":"2022","journal-title":"J. Spacecr. Rockets"},{"key":"mlstaddf10bib4","doi-asserted-by":"publisher","first-page":"889","DOI":"10.1016\/j.vacuum.2011.04.028","article-title":"Numerical assessment of plume heat and mechanical loads and contamination on multi-layer insulation in hard vacuum","volume":"86","author":"Markelov","year":"2012","journal-title":"Vacuum"},{"key":"mlstaddf10bib5","doi-asserted-by":"publisher","first-page":"27","DOI":"10.1016\/j.cja.2012.12.016","article-title":"Research on vacuum plume and its effects","volume":"26","author":"He","year":"2013","journal-title":"Chin. J. Aeronaut."},{"key":"mlstaddf10bib6","doi-asserted-by":"publisher","first-page":"353","DOI":"10.1016\/j.cpc.2004.06.047","article-title":"Simulation of plasma plume-to-spacecraft interaction","volume":"164","author":"Korsun","year":"2004","journal-title":"Comput. Phys. Commun."},{"key":"mlstaddf10bib7","doi-asserted-by":"publisher","DOI":"10.1063\/1.5108534","article-title":"Direct simulation Monte Carlo on petaflop supercomputers and beyond","volume":"31","author":"Plimpton","year":"2019","journal-title":"Phys. Fluids"},{"key":"mlstaddf10bib8","doi-asserted-by":"publisher","DOI":"10.1063\/1.5099042","article-title":"On the basic concepts of the direct simulation Monte Carlo method","volume":"31","author":"Stefanov","year":"2019","journal-title":"Phys. Fluids"},{"key":"mlstaddf10bib9","doi-asserted-by":"publisher","DOI":"10.1063\/5.0051969","article-title":"DSMC study of hypersonic rarefied flow using the cercignani\u2013lampis\u2013lord model and a molecular-dynamics-based scattering database","volume":"33","author":"Liu","year":"2021","journal-title":"Phys. Fluids"},{"key":"mlstaddf10bib10","doi-asserted-by":"publisher","DOI":"10.1063\/5.0045056","article-title":"Effects of the boundary-layer thickness at the cavity entrance on rarefied hypersonic flows over a rectangular cavity","volume":"33","author":"Jin","year":"2021","journal-title":"Phys. Fluids"},{"key":"mlstaddf10bib11","doi-asserted-by":"publisher","first-page":"315","DOI":"10.3390\/aerospace11040315","article-title":"Aerodynamic analysis of deorbit drag sail for cubesat using DSMC method","volume":"11","author":"Chen","year":"2024","journal-title":"Aerospace"},{"key":"mlstaddf10bib12","doi-asserted-by":"publisher","first-page":"565","DOI":"10.1017\/aer.2021.83","article-title":"Simulation of the orbital decay of a spacecraft in low earth orbit due to aerodynamic drag","volume":"126","author":"Kumar","year":"2022","journal-title":"Aeronaut. J."},{"key":"mlstaddf10bib13","doi-asserted-by":"publisher","DOI":"10.1063\/5.0187474","article-title":"Kinetic comparative study of rarefied-transitional flows","volume":"2996","author":"Li","year":"2024","journal-title":"AIP Conf. Proc."},{"key":"mlstaddf10bib14","doi-asserted-by":"publisher","first-page":"4037","DOI":"10.2514\/1.J059333","article-title":"Hypersonic aerodynamics of blunt plates in near-continuum regime by improved navier\u2013stokes model","volume":"58","author":"Ou","year":"2020","journal-title":"AIAA J."},{"key":"mlstaddf10bib15","doi-asserted-by":"publisher","DOI":"10.1016\/j.cpc.2022.108327","article-title":"ADBSat: verification and validation of a novel panel method for quick aerodynamic analysis of satellites","volume":"275","author":"Sinpetru","year":"2022","journal-title":"Comput. Phys. Commun."},{"key":"mlstaddf10bib16","doi-asserted-by":"publisher","first-page":"660","DOI":"10.2514\/1.A34882","article-title":"Hybrid continuum\/rarefied flow simulations of plume interaction with full-sized spacecraft","volume":"59","author":"Tani","year":"2022","journal-title":"J. Spacecr. Rockets"},{"key":"mlstaddf10bib17","doi-asserted-by":"publisher","DOI":"10.1063\/5.0160296","article-title":"Flow time history representation and reconstruction based on machine learning","volume":"35","author":"Zhan","year":"2023","journal-title":"Phys. Fluids"},{"key":"mlstaddf10bib18","doi-asserted-by":"publisher","DOI":"10.1063\/5.0160561","article-title":"Spectral domain graph convolutional deep neural network for predicting unsteady and nonlinear flows","volume":"35","author":"Wen","year":"2023","journal-title":"Phys. Fluids"},{"key":"mlstaddf10bib19","doi-asserted-by":"publisher","DOI":"10.1063\/5.0139907","article-title":"Deep learning predictions of unsteady aerodynamic loads on an airfoil model pitched over the entire operating range","volume":"35","author":"Mohamed","year":"2023","journal-title":"Phys. Fluids"},{"key":"mlstaddf10bib20","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1155\/2022\/9873112","article-title":"visual explainable convolutional neural network for aerodynamic coefficient prediction","volume":"2022","author":"Zhao","year":"2022","journal-title":"Int. J. Aerosp. Eng."},{"key":"mlstaddf10bib21","doi-asserted-by":"publisher","DOI":"10.1063\/5.0155383","article-title":"Prediction of transonic flow over supercritical airfoils using geometric-encoding and deep-learning strategies","volume":"35","author":"Deng","year":"2023","journal-title":"Phys. Fluids"},{"key":"mlstaddf10bib22","article-title":"A deep learning based approach to reduced order modeling for turbulent flow control using LSTM neural networks","author":"Mohan","year":"2018"},{"key":"mlstaddf10bib23","doi-asserted-by":"publisher","DOI":"10.1063\/1.5140772","article-title":"Deep learning methods for super-resolution reconstruction of turbulent flows","volume":"32","author":"Liu","year":"2020","journal-title":"Phys. Fluids"},{"key":"mlstaddf10bib24","doi-asserted-by":"publisher","DOI":"10.1063\/5.0039845","article-title":"Convolutional neural network and long short-term memory based reduced order surrogate for minimal turbulent channel flow","volume":"33","author":"Nakamura","year":"2021","journal-title":"Phys. Fluids"},{"key":"mlstaddf10bib25","doi-asserted-by":"publisher","DOI":"10.1063\/5.0125374","article-title":"Estimating forces from cross-sectional data in the wake of flows past a plate using theoretical and data-driven models","volume":"34","author":"Tong","year":"2022","journal-title":"Phys. Fluids"},{"key":"mlstaddf10bib26","doi-asserted-by":"publisher","DOI":"10.1063\/1.5094943","article-title":"Fast flow field prediction over airfoils using deep learning approach","volume":"31","author":"Sekar","year":"2019","journal-title":"Phys. Fluids"},{"key":"mlstaddf10bib27","doi-asserted-by":"publisher","DOI":"10.1016\/j.ast.2020.105949","article-title":"Fast pressure distribution prediction of airfoils using deep learning","volume":"105","author":"Hui","year":"2020","journal-title":"Aerosp. Sci. Technol."},{"key":"mlstaddf10bib28","first-page":"1","article-title":"CFDNet: a deep learning-based accelerator for fluid simulations","author":"Obiols-Sales","year":"2020"},{"key":"mlstaddf10bib29","doi-asserted-by":"publisher","DOI":"10.1063\/5.0172437","article-title":"A deep learning framework for aerodynamic pressure prediction on general three-dimensional configurations","volume":"35","author":"Shen","year":"2023","journal-title":"Phys. Fluids"},{"key":"mlstaddf10bib30","doi-asserted-by":"publisher","DOI":"10.1063\/5.0033376","article-title":"A point-cloud deep learning framework for prediction of fluid flow fields on irregular geometries","volume":"33","author":"Kashefi","year":"2021","journal-title":"Phys. Fluids"},{"key":"mlstaddf10bib31","doi-asserted-by":"publisher","DOI":"10.1063\/5.0161114","article-title":"Physics-informed graph convolutional neural network for modeling fluid flow and heat convection","volume":"35","author":"Peng","year":"2023","journal-title":"Phys. Fluids"},{"key":"mlstaddf10bib32","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2023.112462","article-title":"Multifidelity deep operator networks for data-driven and physics-informed problems","volume":"493","author":"Howard","year":"2023","journal-title":"J. Comput. Phys."},{"key":"mlstaddf10bib33","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2020.110079","article-title":"PhyGeoNet: physics-informed geometry-adaptive convolutional neural networks for solving parameterized steady-state PDEs on irregular domain","volume":"428","author":"Gao","year":"2021","journal-title":"J. Comput. Phys."},{"key":"mlstaddf10bib34","doi-asserted-by":"publisher","first-page":"A18","DOI":"10.1017\/jfm.2019.814","article-title":"Prediction of turbulent heat transfer using convolutional neural networks","volume":"882","author":"Kim","year":"2020","journal-title":"J. Fluid Mech."},{"key":"mlstaddf10bib35","doi-asserted-by":"publisher","DOI":"10.1063\/5.0165035","article-title":"Nonlinear constitutive calculation method of rarefied flow based on deep convolution neural networks","volume":"35","author":"Yao","year":"2023","journal-title":"Phys. Fluids"},{"key":"mlstaddf10bib36","doi-asserted-by":"publisher","first-page":"1384","DOI":"10.3390\/math9121384","article-title":"Acceleration of Boltzmann collision integral calculation using machine learning","volume":"9","author":"Holloway","year":"2021","journal-title":"Mathematics"},{"key":"mlstaddf10bib37","doi-asserted-by":"publisher","DOI":"10.1016\/j.ast.2022.107852","article-title":"Fast vacuum plume prediction using a convolutional neural networks-based direct simulation Monte Carlo method","volume":"129","author":"Cai","year":"2022","journal-title":"Aerosp. Sci. Technol."},{"key":"mlstaddf10bib38","doi-asserted-by":"publisher","first-page":"269","DOI":"10.1016\/j.cja.2013.02.003","article-title":"Plume aerodynamic effects of cushion engine in lunar landing","volume":"26","author":"He","year":"2013","journal-title":"Chin. J. Aeronaut."},{"key":"mlstaddf10bib39","doi-asserted-by":"publisher","first-page":"67","DOI":"10.1038\/s41592-018-0261-2","article-title":"U-Net: deep learning for cell counting, detection, and morphometry","volume":"16","author":"Falk","year":"2019","journal-title":"Nat. Methods"},{"key":"mlstaddf10bib40","doi-asserted-by":"publisher","first-page":"1856","DOI":"10.1109\/TMI.2019.2959609","article-title":"UNet++: redesigning skip connections to exploit multiscale features in image segmentation","volume":"39","author":"Zhou","year":"2020","journal-title":"IEEE Trans. Med. Imaging"},{"key":"mlstaddf10bib41","doi-asserted-by":"publisher","first-page":"841","DOI":"10.2514\/1.A34026","article-title":"Denoising of direct simulation monte carlo data using proper orthogonal decomposition technique","volume":"55","author":"Kumar","year":"2018","journal-title":"J. Spacecr. Rockets"},{"key":"mlstaddf10bib42","doi-asserted-by":"publisher","DOI":"10.1063\/1.5119595","article-title":"Denoising quasi-steady state particle simulations","volume":"2132","author":"Martin","year":"2019","journal-title":"AIP Conf. Proc."},{"key":"mlstaddf10bib43","doi-asserted-by":"publisher","first-page":"20072","DOI":"10.1038\/s41598-022-24226-1","article-title":"Efficient two-stage detection of human-object interactions with a novel unary-pairwise transformer","author":"Zhang","year":"2022"},{"key":"mlstaddf10bib44","article-title":"Transformer for partial differential equations\u2019 operator learning","author":"Li","year":"2023"},{"key":"mlstaddf10bib45","article-title":"Transolver: a fast transformer solver for pdes on general geometries","author":"Wu","year":"2024"}],"container-title":["Machine Learning: Science and Technology"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/addf10","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/addf10\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/addf10","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/addf10\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/addf10\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/addf10\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/addf10\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/addf10\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,12]],"date-time":"2025-06-12T07:17:21Z","timestamp":1749712641000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2632-2153\/addf10"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,12]]},"references-count":45,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2025,6,12]]},"published-print":{"date-parts":[[2025,6,30]]}},"URL":"https:\/\/doi.org\/10.1088\/2632-2153\/addf10","relation":{},"ISSN":["2632-2153"],"issn-type":[{"value":"2632-2153","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6,12]]},"assertion":[{"value":"Efficient prediction of aerodynamic forces in rarefied flow using convolutional neural network based multi-process method","name":"article_title","label":"Article Title"},{"value":"Machine Learning: Science and Technology","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2025 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2024-08-30","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-05-30","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2025-06-12","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}