[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120063-en":3,"doc-seo-120063-105":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},120063,5909877438554,"Maeve","https://ap-avatar.wpscdn.com/avatar/5600025385ad2bf12a7?_k=1778553567797529272",8,"Research & Report","A machine learning study to predict wind-driven water runback characteristics","The unsteady runback behavior of wind-driven runback water film (WDRWF) flows over aircraft surfaces strongly influences aircraft icing, a major aviation hazard in cold weather. Limited insight into multiphase interactions among free-stream airflow, water-film motion, and the solid airframe surface limits conventional theoretical and numerical methods from accurately simulating WDRWF flow. A machine learning framework predicts the front contact point evolution and the film thickness distribution, comparing LightGBM and Multi-Layer Perceptron for front contact point and using a ConvLSTM-AutoEncoder for spatial-temporal thickness forecasting, including robustness to noisy inputs and evaluation on unseen datasets.","Physics of Fluids ARTICLE  \n[pubs.aip.org/aip/pof](pubs.aip.org/aip/pof)  \nA machine learning study to predict wind-driven water runback characteristics  \n\n| Cite as: Phys. Fluids 35, 102104 (2023); doi: 10.1063/5.0167545 Submitted: 13 July 2023 . Accepted: 15 September 2023 .\u003Cbr>Published Online: 3 October 2023 |  |  |  |\n| --- | --- | --- | --- |\n| Jincheng Wang,  Haiyang Hu,  Ping He,  and Hui Hua)  |  |  |  |\n| AFFILIATIONS\u003Cbr>Department of Aerospace Engineering, Iowa State University, Ames, Iowa 50011-1096, USA\u003Cbr>a)Author to whom correspondence should be addressed: [huhui@iastate.edu](huhui@iastate.edu) |  |  |  |\n| ABSTRACT\u003Cbr>The unsteady runback behavior of wind-driven runback water film (WDRWF) flows over aircraft surfaces has a significant impact on the aircraft icing process, one of the most significant aviation hazards in cold weather. The limited understanding of the complex multiphase interactions between freestream airflow, water film motion, and solid airframe surface makes conventional theoretical/numerical methods unable to precisely simulate WDRWF flow. Machine learning-based techniques can accurately capture complex physics using data, making it an attractive alternative to conventional methods. In this study, machine learning methods are used to predict the evolution of the front contact point (FCP) of WDRWF flow and film thickness distribution (FTD) of WDRWF flow. For FCP prediction, the performance of the Light Gradient-Boosting Machine (LightGBM) and Multi-Layer Perceptron is compared quantitatively. They perform well in capturing intermittent and smooth features, respectively. For the prediction of the spatial-temporal evolution of FTD, a computationally efficient deep neural network architecture named ConvLSTM-AutoEncoder was developed, which predicts a future FTD based on a sequence of FTDs in the past. The robustness of the ConvLSTM-AutoEncoder model to noisy input FTD is demonstrated. The generalizability of the three models is evaluated by applying the trained models to unexplored datasets. Based on the proposed techniques’ generalizability, robustness, and computational efficiency, machine learning-based methods are demonstrated to be powerful tools in predicting the complex unsteady characteristics of the multiphase WDRWF flows.\u003Cbr>Published under an exclusive license by AIP Publishing. [https://doi.org/10.1063/5.0167545](https://doi.org/10.1063/5.0167545) |  |  |  |\n\nI. INTRODUCTION  \nAircraft icing is a well-known weather hazard affecting aircraft performance and flight safety greatly. As ice structures accrete over airframe surfaces, the aerodynamic performance of an airplane would degrade significantly with the aerodynamic drag increasing dramatically and lift force decreasing rapidly. Ice accretion can usually be categorized into rime and glaze icing. When the ambient temperature is low enough (e.g., less than 􀀂8 􀀃 C) under a relatively dry condition, the supercooled water droplets would be frozen instantly to form rime ice upon the impingement onto the airframe surface. 1,2 Glaze icing usually occurs under conditions with relatively warmer ambient temperatures (i.e., just below the freezing point of water), higher liquid water content (LWC), and larger supercooled water droplets. In a glaze ice accretion process, only a portion of the supercooled water droplet would be frozen into solid ice upon impact with the airframe surfaces of an airplane, while the rest of the impacted water droplets would stay in liquid and move freely over the airframe surfaces in the form of water film/rivulet flows as driven by the boundary layer airflow.3–5 The wind-driven runback characteristics of the unfrozen water film/rivulet  \nflows would directly or indirectly affect the ice formation and accretion over the airframe surfaces.6 The transient behavior of unfrozen runback water flow would directly affect the transportation process of impinged water mass over the ice-accreting airframe surfaces.7 Local convective heat transfer a","cbCail55RQASvpQT","https://ap.wps.com/l/cbCail55RQASvpQT","pdf",8474635,1,17,"English","en",105,"# Abstract\n# I. INTRODUCTION\n## Aircraft icing background\n## Wind-driven runback water film relevance\n## Motivation for machine learning","[{\"question\":\"Why is predicting wind-driven water runback characteristics important for aircraft icing?\",\"answer\":\"Wind-driven runback water film and rivulet flows affect ice formation and accretion by influencing the transport of impinged water mass and local convective heat transfer, which changes icing rates and structures.\"},{\"question\":\"How does the study predict the front contact point (FCP) of WDRWF flow?\",\"answer\":\"It quantitatively compares LightGBM and a Multi-Layer Perceptron for FCP prediction, showing that they capture intermittent and smooth features respectively.\"},{\"question\":\"What model is used for forecasting the film thickness distribution (FTD), and how is robustness evaluated?\",\"answer\":\"A ConvLSTM-AutoEncoder architecture is developed to predict future FTD from past sequences. Robustness is demonstrated by testing the model against noisy input FTD and evaluating generalizability on unexplored datasets.\"}]","A machine learning study to predict wind-driven water runback characteristics | PDF",1785727949,43,{"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},"a-machine-learning-study-to-predict-wind-driven-water-runback-characteristics","",{"@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/a-machine-learning-study-to-predict-wind-driven-water-runback-characteristics/120063/",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},"Why is predicting wind-driven water runback characteristics important for aircraft icing?","Question",{"text":75,"@type":76},"Wind-driven runback water film and rivulet flows affect ice formation and accretion by influencing the transport of impinged water mass and local convective heat transfer, which changes icing rates and structures.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the study predict the front contact point (FCP) of WDRWF flow?",{"text":80,"@type":76},"It quantitatively compares LightGBM and a Multi-Layer Perceptron for FCP prediction, showing that they capture intermittent and smooth features respectively.",{"name":82,"@type":73,"acceptedAnswer":83},"What model is used for forecasting the film thickness distribution (FTD), and how is robustness evaluated?",{"text":84,"@type":76},"A ConvLSTM-AutoEncoder architecture is developed to predict future FTD from past sequences. Robustness is demonstrated by testing the model against noisy input FTD and evaluating generalizability on unexplored datasets.","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"]