空间齿轮系统微动载荷谱自适应提取及疲劳寿命预测方法研究

Fatigue damage modeling and life prediction are the key technologies to ensure the safe and reliable operation of engineering equipment. The micro-vibration mechanical behavior of the space gear is complex and its characteristics are weak, which makes it difficult to extract the load spectrum. However, due to the severe experimental conditions of the space gear, the fretting fatigue life prediction of the gear in the space micro-vibration environment is even more difficult. The research on space gear fretting load extraction and life prediction methods, which are mainly oriented to the ground design stage, is difficult to apply under complex space conditions. Therefore, how to realize space gear damage modeling and fatigue life prediction in micro-vibration environment, and then realize space gear anti-fatigue design is a frontier scientific problem that needs to be solved. To this end, the project conducts research on fretting damage modeling and fatigue life prediction methods for space gear systems. The main contents include: 1) adaptive extraction of micro-vibration load spectrum; 2) fretting damage modeling and fatigue life prediction; 3) anti-fatigue design of the gear system based on fretting damage modeling and life prediction. The project focuses on the adaptive extraction of harmonic interference load spectrum under micro-vibration conditions and the fatigue life prediction under the influence of multi-parameters, and reveals the quantitative relationship between the micro-motion characteristic parameters and fatigue life, and improves the space gear system through anti-fatigue design. The above research results provide basic theory and key technologies for the prediction of fretting fatigue life of space gears.

疲劳损伤建模及寿命预测是保障工程设备安全可靠运行的关键技术。空间齿轮微振动力学行为复杂、特征微弱，导致其载荷谱难以提取，而受空间齿轮苛刻实验条件限制，空间微振动环境下齿轮微动疲劳寿命预测更是难上加难。现有研究主要面向地面设计阶段的空间齿轮微动载荷提取及寿命预测方法在空间复杂条件下难以适用。因此，如何实现微振动环境下空间齿轮损伤建模及疲劳寿命预测，进而实现空间齿轮抗疲劳设计，是亟需解决的前沿科学问题。为此，项目面向空间齿轮系统开展微动损伤建模及疲劳寿命预测方法研究，主要内容包括：1）微振动载荷谱自适应提取；2）微动损伤建模及疲劳寿命预测；3）基于微动损伤建模及寿命预测的抗疲劳设计。重点解决微振动条件下谐波干扰载荷谱的自适应提取、多参量影响下疲劳寿命预测等关键问题，揭示微动特征参量与疲劳寿命的定量关系，通过抗疲劳设计提高空间齿轮系统的可靠性，为空间齿轮的微动疲劳寿命预测提供基础理论与关键技术。

齿轮系统作为传递运动和动力的关键机构，是目前各航天器中应用最为广泛的传动部件之一，随着航天任务复杂程度和工作时限的逐步增加，对空间装备零部件的可靠性和工作寿命的要求越来越高。高效准确地监测齿轮和轴承等空间活动部件的健康状态对于保障空间装备服役安全十分关键。本项目研究主要是为克服微振动条件下空间齿轮系统的载荷谱分析及其寿命预测难题：1）微振动条件下齿轮载荷谱自适应提取的抗噪性以及对先验知识依赖性；2）微振动条件下齿轮在线健康评估及其寿命预测。.针对微振动下齿轮载荷谱分析和特征提取的难题，分别建立基于熵理论的载荷谱特征评估方法、基于模态分解理论的载荷特征提取方法和基于解卷积理论的载荷谱特征提取方法用于齿轮故障特征相关信息的识别和提取，所提方法在抗噪性以及对先验知识依赖性都得到了很大程度的改善。.针对微振动条件下齿轮健康评估及其寿命预测问题，分别提出了基于机器学习齿轮健康状态评估方法、基于深度学习的齿轮剩余寿命预测方法和基于维纳过程的齿轮剩余寿命预测方法等，实现了复杂条件下齿轮等关键旋转机械的故障诊断和在线剩余寿命预测。.基于本项目相关研究成果，标注基金号的学术论文共发表25篇（一作/通讯论文19篇），其中SCI期刊论文22篇，ESI高被引论文5篇，授权发明专利6项，发表在《Mechanical Systems and Signal Processing》和《Structural Health Monitoring-An International Journal》的2篇第一作者且本基金为第一标注的论文都被引用超过60次，相关研究成果得到了清华大学、杰青褚福磊教授，北京航空航天大学、杰青/长江学者林京教授，西安交通大学、杰青陈雪峰教授，西安交通大学大学、杰青雷亚国教授，美国伊利诺伊大学Paolo Gardoni教授等著名专家学者引用评价。