压力容器2025，Vol.42Issue(7)：22-29,8.DOI:10.3969/j.issn.1001-4837.2025.07.003
基于机器学习的管束流体弹性不稳定性预测方法研究
Research on prediction method for fluid-elastic instability of tube bundles based on machine learning
摘要
Abstract
To address the safety issues of shell-and-tube heat exchangers caused by fluid-elastic instability(FEI)due to flow-induced vibration,a two-phase water tunnel test setup was built to investigate the effect of void fraction on the fluid-elastic instability characteristics of a single flexible tube with unequal natural frequencies in rigid tube arrays with different pitch-to-diameter ratios(P/D=1.41,1.483).A machine learning algorithm was introduced to construct a support vector machine model for predicting fluid-elastic instability,obtaining the critical flow velocity for fluid-elastic instability and verifying the model's stability and accuracy.The results show that the prediction results using the machine learning algorithm based on the support vector machine model have errors 19%lower than those of the quasi-static model,14.2%lower than those of the unsteady model,and 14.5%lower than those of the quasi-steady model.The research results can provide references for predicting fluid-elastic instability in heat exchangers.关键词
管壳式换热器/流体弹性不稳定性/机器学习Key words
shell-and-tube heat exchanger/fluid-elastic instability/machine learning分类
机械制造引用本文复制引用
卢玲玲,孙正瑞,朱国瑞..基于机器学习的管束流体弹性不稳定性预测方法研究[J].压力容器,2025,42(7):22-29,8.基金项目
中国广核集团尖峰计划项目(3100173757) （3100173757）
