摘要
针对航空发动机涡轮盘蠕变疲劳寿命预测计算效率低、实时性差、难以满足工程实际中寿命管理需求的问题,本文提出了涡轮盘蠕变疲劳寿命预测的数字孪生模型。该模型首先通过降阶方法获取涡轮盘的应力场,然后根据寿命预测模型计算涡轮盘的蠕变疲劳寿命。采用降阶方法计算应力的优势是可以快速得到不同工况下的结构应力场,相较于传统有限元分析,应力场计算时间减少了99.74%,可实现秒级响应。基于应力场计算结果,模型可实时预测涡轮盘在不同工况下的蠕变疲劳寿命,从而解决了涡轮盘蠕变疲劳寿命预测实时性差的问题。为验证模型可靠性,开展了6组不同工况条件下涡轮盘榫槽底部模拟件的蠕变疲劳试验,结果表明:模型预测寿命与试验寿命的相对误差均能控制在1.5倍误差带范围内,证明该模型具有很高的准确性。
The computational efficiency and real-time performance of creep-fatigue life prediction for aerospace engine turbine disks are low,making it difficult to meet the requirements of life management in practical engineering applications.Therefore,a digital twin model for predicting the creep-fatigue life of turbine disks is proposed.The model employs reduced-order model(ROM)methods to rapidly derive structural stress fields under varying operating conditions,followed by life prediction based on stress-driven fatigue-creep damage models.First,a parametric ROM is constructed to replace high-fidelity finite element analysis(FEA)for stress field computation.By projecting the fullorder governing equations onto a low-dimensional subspace using techniques such as singular value decomposition(SVD),the ROM achieves a 99.74%reduction in computational time compared to conventional FEA—enabling stress field predictions within seconds under varying operational conditions.Leveraging these real-time stress results,the model realizes instantaneous prediction of creep-fatigue life across diverse operational scenarios,effectively resolving the latency issues of conventional methods.To validate the reliability of the model,six groups of creep-fatigue tests were conducted on specimens corresponding to the turbine disk bottom under varying conditions.Results demonstrate that the relative errors between predicted and experimental lifespans consistently fall within a 1.5-fold error band,confirming the model’s high prediction accuracy.The proposed framework significantly enhances computational efficiency while maintaining precision,offering a practical solution for real-time lifespan monitoring and management of turbine disks in engineering applications.This advancement bridges the gap between theoretical models and industrial requirements,providing a robust foundation for proactive maintenance strategies in aviation systems.
作者
孙雨杰
朱郁涵
康磊
张润泽
谈建平
温建锋
刘长利
SUN Yujie;ZHU Yuhan;KANG Lei;ZHANG Runze;TAN Jianping;WEN Jianfeng;LIU Changli(School of Mechanical and Power Engineering,East China University of Science and Technology,Shanghai 200237,China)
出处
《华东理工大学学报(自然科学版)》
北大核心
2025年第6期857-867,共11页
Journal of East China University of Science and Technology
基金
国家自然科学基金(52075176)。
关键词
蠕变疲劳
寿命预测
降阶模型
数字孪生
涡轮盘
creep fatigue
life prediction
reduced order model
digital twin
turbine disk
