摘要
航空发动机是飞机的心脏,其复杂结构以及恶劣运行工况致使故障频发,对高可靠性的需求日益增长.这不仅对飞行安全构成挑战,同时也导致了高额的维护成本.为提升航空发动机可靠性并控制维护成本,故障诊断与健康管理技术至关重要.本文重点探讨航空发动机健康管理中叶端定时感知、振动快变信号分析,以及智能检测与健康监测系统等多项关键技术的研究进展.考虑到传统转子叶片监测面临传递路径长、振动特征微弱的问题,本文介绍适用于转子叶片直接监测的叶端定时感知技术,并考虑到航发监测中少传感、低介入需求,提出了每级1~2支传感的低介入叶端定时测振方法.考虑到航空发动机快变特征的提取和处理已超出现有诊断方法的能力,本文介绍了适用于快变信号分析的瞬时频带量化分解方法,并通过航空发动机整机试车数据验证,说明其在中介轴承诊断与振动突跳溯源中的有效性.考虑多变飞行状态场景下HUMS系统面临的技术挑战,本文介绍了变分解耦异常检测技术用于提取飞行状态无关的监测指标,并通过右发输入轴真实飞行故障案例分析了其在传动系统异常检测中的有效性.此外,本文还对可解释人工智能、机器人检测等新兴技术的发展进行前瞻性展望,共同构建发动机全面、实时的健康监测和故障预警体系,为飞行安全提供更为坚实的保障.
Aero-engine is the heart of the aircraft.Its complex structure and poor operating conditions cause frequent failures,and the demand for high reliability is increasing.This not only poses a challenge to flight safety,but also leads to high maintenance costs.Aero-engine fault prediction and health management technology is a key way to improve operational reliability and minimize maintenance requirements.To establish a complete aero-engine health management system,this paper reviews advancements in several key technologies,such as blade tip timing technology,vibration superfast signal processing technology,intelligent detection and health and usage monitoring systems(HUMS).To overcome the limitation of long transmission paths and weak vibration response in traditional rotating blade monitoring,this paper introduces blade tip timing technique for direct blade vibration measurement.In addition,to satisfy the few-probe and low-intervention requirement in aero-engine health monitoring,we propose two efficient blade tip timing methods that rely on only 1 or 2 probes.Considering that the superfast signal from aero-engines have exceeded the limitations of existing diagnostic methods,this paper proposes instantaneous frequency bands and synchronized compression transformation for analyzing these signals.The method’s effectiveness in the diagnosis of intermediate bearings and the tracing of vibration jumps is verified through the whole-engine test data of aero-engines.Aiming at the technical challenges faced by the HUMS system under variable flight conditions,this paper introduces the latest variable decoupling anomaly detection technology for extracting flight state-independent monitoring indicators,and analyzes its effectiveness in the HUMS system through a real flight fault case of the right engine shaft.Furthermore,this paper discusses the potential of emerging technologies such as interpretable artificial intelligence and robot detection.By integrating these advancements,we aim to construct a comprehensive,real-time health monitoring,and fault warning system that enhances flight safety.
作者
陈雪峰
王诗彬
杨志勃
孙闯
丁宝庆
CHEN XueFeng;WANG ShiBin;YANG ZhiBo;SUN Chuang;DING BaoQing(National Key Lab of Aerospace Power System and Plasma Technology,Xi’an Jiaotong University,Xi’an 710049,China;School of Mechanical Engineering,Xi’an Jiaotong University,Xi’an 710049,China)
出处
《中国科学:技术科学》
北大核心
2025年第8期1255-1280,共26页
Scientia Sinica(Technologica)
基金
国家自然科学基金(批准号:92060302,92360305,92360306)资助项目。
