Course Hero Logo

Paper_International.pdf - Graphical Abstract Hybrid...

Course Hero uses AI to attempt to automatically extract content from documents to surface to you and others so you can study better, e.g., in search results, to enrich docs, and more. This preview shows page 1 - 4 out of 6 pages.

Graphical AbstractHybrid critical heat flux prediction method in TRIGA-fueled reactors cooled by natural convectionusing machine learning modelBambang Riyono,Anhar Riza Antariksawan,Muhammad Reza Pulungan,Andi Dharmawan
HighlightsHybrid critical heat flux prediction method in TRIGA-fueled reactors cooled by natural convectionusing machine learning modelBambang Riyono,Anhar Riza Antariksawan,Muhammad Reza Pulungan,Andi DharmawanResearch highlights item 1Research highlights item 2Research highlights item 3
Hybrid critical heat flux prediction method in TRIGA-fueled reactorscooled by natural convection using machine learning model?,??BambangRiyonoa,c,<,1(Researcher),Anhar RizaAntariksawan (Researcher),MuhammadRezaPulunganb,c,2(Co-ordinator) andAndiDharmawana,c,<<,1,3aElsevier B.V., Radarweg 29, 1043 NX Amsterdam, The NetherlandsbSayahna Foundation, Jagathy, Trivandrum 695014, IndiacSTM Document Engineering Pvt Ltd., Mepukada, Malayinkil, Trivandrum 695571, IndiaA R T I C L E I N F OKeywords:hybrid methodsteady state experimentalmachine learningCHFpredictionA B S T R A C TCritical Heat Flux (CHF) plays importance roles for nuclear reactor safety because when it happenleads the ability of the coolant to remove heat from the cladding surface is greatly reduced, and thiscan lead to sharply increased fuel temperatures and potential fuel failure. Prediction of CHF could beprocessed by using some code or computer program, such as SPACE, TRACE, and PARET or usingLook Up Table (LUT) Method. Unfortunately, the problem when the using of these code to predictCHF are necessity rigidly iterations and consumes a large amount of computing time. Moreover, theproblem when the using of Look Up Table Method is necessity complex correction factor because ofthe lack of measured CHF data in the region of interest to TRIGA reactors, so none of these sets ofdata closely match TRIGA conditions. This paper proposes a solution of the problems by providing asimple and accurate prediction method based on a hybrid method using machine learning. The aim ofthis paper is to provide hybrid method using machine learning to predict CHF with acceptable level ofaccuracy and investigated the factor that contribute during CHF mechanism. This research used hybridmodel that incorporate among SVR, mathematical model and steady state experimental data. Whilemathematical model serves to lay the groundwork and provide a baseline solution, SVR is used to learnfrom the residual between steady state experimental data and mathematical model predicted output(not from the final output/target). The developed SVR models were implemented to extrapolate valueof CHF predicted which it is a blind case or untrained data. The accuracy of prediction was investigatedbased on comparison with prediction using existing CHF correlation and LUT method. This resultsshow that proposed hybrid method have good accuracy and closely agreement with calculation codeand LUT1. Introduction

Upload your study docs or become a

Course Hero member to access this document

Upload your study docs or become a

Course Hero member to access this document

End of preview. Want to read all 6 pages?

Upload your study docs or become a

Course Hero member to access this document

Term
Fall
Professor
NoProfessor

Newly uploaded documents

Show More

Newly uploaded documents

Show More

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture

  • Left Quote Icon

    Student Picture