logo
CanyamAI summaries for academic research
language
MobileMobile
  • Phone FeatureWeChat Mini Program
EnglishEnglish
noticeNotice
BlogBlog
delete
arrow
Return

Predicting Pressure Loss in Circular Pipe using AI-Aided Engineering

delete2026-03-01
delete0
PRE
AI
L
Lee, Jong Hui *
L
Lee, Gyeongseop
O
Oh, Jin Ho
R
Ryu, Seungyeob
H
Han, Hun Sik
A
Ahn, Kwanghyun
DOI：10.3795/KSME-B.2026.50.3.157delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
Since NASA introduced NASTRAN in the 1960s, computer-aided engineering technologies have evolved significantly through advances in finite element analysis and computational fluid dynamics. Recently, the emergence of artificial intelligence (AI) has spurred growing interest in AI-aided engineering (AAE). In this study, a multilayer perceptron and a large language model were used to predict the friction factor in circular pipes, which is a key parameter in the iterative design of flow channels in the internal structures of reactors. Both AAE models exhibited the capability to predict the friction factor accurately across the entire flow regime, including laminar, transitional, and turbulent flow, with errors within 10 % relative to the ground-truth values. With further refinement and extension, the proposed AAE models have the potential to serve as effective tools to support repetitive design iterations in reactor engineering.
Keywords:
Artificial Intelligence
Pressure Loss
Circular Pipe
Artificial Neural Network
ChatGPT

Journal

T
Transactions of the Korean Society of Mechanical Engineers B
IF:
0.2
Papers:
37
Citations:
0

Organization

No organization information available