{"id":16440,"date":"2024-04-25T10:17:47","date_gmt":"2024-04-25T10:17:47","guid":{"rendered":"https:\/\/aerospacerepository.org\/?p=16440"},"modified":"2024-04-26T15:00:13","modified_gmt":"2024-04-26T15:00:13","slug":"surrogate-modelling-of-supersonic-fuel-air-mixing-in-a-multi-strut-injectionscramjet-engine-using-artificial-neural-network","status":"publish","type":"post","link":"https:\/\/aerospacerepository.org\/index.php\/2024\/04\/25\/surrogate-modelling-of-supersonic-fuel-air-mixing-in-a-multi-strut-injectionscramjet-engine-using-artificial-neural-network\/","title":{"rendered":"Surrogate modelling of supersonic fuel-air mixing in a multi-strut injectionscramjet engine using artificial neural network"},"content":{"rendered":"\n

Ali Can Ispir, Kamila Zdyba\u0142, Bayindir H. Saracoglu, Thierry Magin, Axel Coussement<\/strong><\/p>\n\n\n\n

DOI Number XXX-YYY-ZZZ<\/strong><\/p>\n\n\n\n

Conference Number HiSST-2022-425<\/strong><\/p>\n\n\n\n

One of the main challenges which must be hurdled in the dual mode ramjet\/scramjet engines\u2019 design pro-
cess, is to obtain an optimal fuel-air mixing distribution and a good penetration. One way is to enhance
the mixing efficiency is to inject the fuel via multiple struts in parallel direction and make the fuel be mixed
with air throughout the shock-expansion waves structure. Mixing intensity is augmented by the turbu-
lence level highly depends on the strut geometrical parameters, however, this enhancement in the mixing
can bring about several detrimental effects on the aerodynamic features of the propulsion system such
as increasing the losses on total pressure. Therefore, optimizing configurations of the fuel struts and
resolving the interaction between the design variables are obligatory in order to yield best overall engine
performance. The present study focuses on the investigation of the strut design parameters impacts on
the fuel-air mixing and aerodynamic properties particularly efficiency, length and total pressure recovery
factor. We solved compressible non-reactive RANS filtered governing equations on the 2D flow domain
of a dual mode ramjet engine (operating in scramjet mode) combustor. Exploring the design space of
the fuel struts in terms of mixing and total pressure losses, requires plenty of simulations and an infor-
mative dataset. Performing the simulations in every single design point is computationally prohibitively
expensive. Machine learning techniques thus can be a key solution for multi-objective optimization of
design variables, making the predictions by utilizing a database having a number of observations and
generating reduced-order models that can be used in the preliminary design exercises. In present work,
we created a CFD database having 100 observation points with three varying design variables: strut lo-
cation, strut wedge angle and strut V-settlement angle. We applied Artificial neural network regression
model to this database in order to formulate the mixing efficiency of multi-strut injection scramjet engine.
We discuss the deep learning model prediction accuracy by computing coefficient of determination, R2
and drawing the parity plots for each objective function. In our findings in the investigation of the flow
physics, the wedge angle is the dictating parameter for the shock-expansion wave structure in the post
strut region and accordingly the mixing and aerodynamic performance of the engine.<\/p>\n\n\n\n

Read the full paper here<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"

Ali Can Ispir, Kamila Zdyba\u0142, Bayindir H. Saracoglu, Thierry Magin, Axel Coussement<\/b><\/p>\n

DOI Number XXX-YYY-ZZZ<\/p>\n

Conference Number HiSST-2022-425<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[993,1006],"tags":[1340,1341,1224,1343,1342,1344],"class_list":["post-16440","post","type-post","status-publish","format-standard","hentry","category-events","category-hisst-2022","tag-high-speed-propulsion","tag-hydrogen-fueled-engine","tag-machine-learning","tag-mixing-efficiency","tag-multi-strut-injection","tag-total-pressure-recovery-factor","category-993","category-1006","description-off"],"_links":{"self":[{"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/posts\/16440","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/comments?post=16440"}],"version-history":[{"count":2,"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/posts\/16440\/revisions"}],"predecessor-version":[{"id":16498,"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/posts\/16440\/revisions\/16498"}],"wp:attachment":[{"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/media?parent=16440"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/categories?post=16440"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/aerospacerepository.org\/index.php\/wp-json\/wp\/v2\/tags?post=16440"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}