一种适用于低压涡轮叶片转捩-分离流预测的极大涡模拟方法

Translated title of the contribution: A very large eddy simulation method for predicting transition and separation flow on low-pressure turbine blades

Yuyang Mu, Jiakuan Xu, Yi Li, Guoyu Zheng, Lei Qiao, Junqiang Bai

Research output: Contribution to journalArticlepeer-review

Abstract

In the domain of aero engine, high adverse pressure gradient effects on the surface of high load low-pressure turbine blades can induce transition-separation flows with highly unsteady flow characteristics. Conventional engineering applications find it challenging to simulate these phenomena using transition models basedon the Reynolds Averaged Navier-Stokes (RANS) framework. To enhance prediction accuracy, an improved approach is proposed based on the Very Large Eddy Simulation (VLES) method, further coupled with the ɤ transition model suitable for turbomachinery flows, resulting in the ɤ-VLES model. This model is utilized to forecastlaminar separation bubble-induced transition and separation flow evolution on the surface of the Pak-B low-pressure turbine blades. The outcomes demonstrate that the ɤ-VLES model accurately computes time-averaged pressure coefficients and velocity profiles on the blade surface, revealing the shedding process of periodic vorticeswithin long separation bubbles’ transient flow fields. The simulation results align closely with experimental data, effectively addressing the shortcomings of the (Formula presented) and ɤ transition models under the RANS framework in predicting transition-separation flows.

Translated title of the contributionA very large eddy simulation method for predicting transition and separation flow on low-pressure turbine blades
Original languageChinese (Traditional)
Article number2402028
JournalTuijin Jishu/Journal of Propulsion Technology
Volume46
Issue number5
DOIs
StatePublished - 1 May 2025

Fingerprint

Dive into the research topics of 'A very large eddy simulation method for predicting transition and separation flow on low-pressure turbine blades'. Together they form a unique fingerprint.

Cite this