摘要
齿轮磨损过程是一系列理化效应的结果,能够直接表征齿轮磨损的直接状态信息磨损量往往是不可测量或难于测量的;另一方面,状态监测系统得到大量反映齿轮健康状态的间接状态信息。针对有限的直接状态信息和大量的间接状态信息问题,结合齿轮箱全寿命实验结果,将振动信号特定频带能量作为间接状态信息,建立了Gamma-状态空间模型,提出了经验最大化算法和粒子滤波相结合的参数求解方法。利用该模型对齿轮箱齿轮磨损情况进行了全面分析,并进行了磨损预测研究,通过与实验结果对比分析,验证了模型的有效性。
The process of gear wear is the result of physical and chemical effect. Amount of indirect condition information is obtained by condition monitoring system. Wear is the dominant reason of the change of indirect condition information. Combined with the gearbox full lifetime experiments, which obtain limited direct condition information and plentiful indirect condition information, the Gamma-State space model is established. Experience maximization arithmetic and particle filtering are used to solve model parameters. Then the model is used to analyze the gear wear and predict the development of wear. Compared with the experimental results, the efficiency of the model is validated.
出处
《机械科学与技术》
CSCD
北大核心
2011年第12期2086-2091,共6页
Mechanical Science and Technology for Aerospace Engineering
基金
总装重点预研基金项目(9140A27020308JB34)资助
