Prediction of blowdown of a pressure relief valve using response surface methodology and CFD techniques

Jian Zhang, Lui Yang, William Dempster, Xinhai YU, Jiuhomg Jia, Shan-Tung Tu

Research output: Contribution to journalArticlepeer-review

29 Citations (Scopus)
501 Downloads (Pure)

Abstract

In this study, parametric assessment of the main geometric design features of a pressure relief valve (PRV) with a backpressure chamber and two adjusting rings was conducted using response surface methodology. This design approach was established by using computational fluid dynamics (CFD) to model the dynamic performance of the opening and closing of a nuclear power main steam pressure relief valve (NPMS PRV). An experimental facility was established to test the NPMS PRV in accordance with the standard ASME PTC 25, and to validate the CFD model. It was found that the model can accurately simulate the dynamic performance of the NPMS PRV; the difference in blowdown between the simulation and experiment results is found to be below 0.6%. Thus, themodel can be used as part of a design analysis tool. The backpressure chamber assisted in the reseating and decreased the blowdown of the NPMS PRV from 18.13% to 5.50%. The sensitivity to valve geometry was investigated, and an explicit relationship between blowdown and valve geometry was established (with a relative error less than 1%) using the response surface methodology; this will allow designers to assess the valve settings without the need for a CFD model.
Original languageEnglish
Pages (from-to)713-726
Number of pages14
JournalApplied Thermal Engineering
Volume133
Early online date31 Jan 2018
DOIs
Publication statusPublished - 25 Mar 2018

Keywords

  • pressure relief valve
  • back pressure chamber
  • response surface methodology
  • transient numerical simulation
  • computational fluid dynamics

Fingerprint

Dive into the research topics of 'Prediction of blowdown of a pressure relief valve using response surface methodology and CFD techniques'. Together they form a unique fingerprint.

Cite this