RV减速器轴承故障的信息多源耦合机制与诊断方法研究

RV reducers as the core part have been equipped widely in industrial robots, CNC machine tools and other high-end precision mechanical equipment. Rolling element bearing is one of the crucial and vulnerable components in the RV reducer. The operating condition of bearings will directly affect the performance of the reducer or even the whole mechanical equipment. Due to the particularity of mechanical structure and movement, compared with the traditional monitoring object, the multi-source coupling mechanism of RV reducer bearing fault is more complex which will result in more difficulty in fault diagnosis research. From the vibration behavior, fault mechanism to signal analysis, at multiple levels the project intends to explore and establish the dynamic model of bearing failure under time-varying coupling excitation. Through the study on the law of fault vibration, the multi-source coupling mechanism is revealed. Based upon this, the essence of the interference and fault components is clarified. The mapping relationship between fault information and quantitative evaluation attributes of indexes is studied to propose the optimal scheme of feature indicators under different filtering strategies. Further research on the customized design of noise suppression scheme according to different fault types, and the strong adaptive minimum spectral entropy deconvolution method for the targeted enhancement of the weak fault characteristics is made, and by which, the RV reducer bearing health monitoring technology is established. The research work in this project could not only guarantee for RV reducer security but also help to improve the holistic maintenance ability of the equipment, which has important theoretical significance and engineering value.

RV减速器作为核心部件已广泛配备在工业机器人、数控机床等高端精密机械装备中。滚动轴承是RV减速器中关键且易损的零部件，其运行状态的好坏将直接影响减速器甚至整机的性能。由于机械结构和运动方式的特殊性，相比传统监测对象，RV减速器轴承故障信息多源耦合机制更为复杂，开展故障诊断研究的难度更大。项目从振动行为、故障机理到信号分析多层次进行探索，拟建立时变耦合激励作用下的轴承故障动力学模型，研究故障振动规律，揭示多源耦合机制；在此基础上，厘清干扰激励与故障成分的本质，研究故障信息与指标定量评估属性的映射关系，提出不同滤波策略下的特征指标优选方案；进一步，针对不同故障定制化设计噪声抑制方案，结合强自适应的最小谱熵解卷积方法对微弱故障特征进行靶向增强，以此建立RV减速器轴承健康监测技术。本项目的研究不仅能保障RV减速器的稳定运行，还能提高设备整体可靠性水平，具有重要的理论意义和工程应用价值。

本项目围绕机器人用RV减速器轴承安全服役与故障诊断中面临的关键科学问题与技术难题开展了RV减速器轴承故障行为表征、故障特征解析与诊断方法等基础理论方面的研究。首先，建立了RV减速器典型故障与内置编码信息时频表征之间的映射关系，在此基础上，提出了基于解卷积、变分模态分解和奇异值分解等理论的内置编码信息处理技术。其次，提出了不同滤波策略下的特征指标优选框架，为指导故障特征提取方案奠定基础，并率先引入经济学的基尼指数用于机械故障特征评价，由此建立了基于基尼指数的轴承故障诊断方法。另外，针对轴承的不同故障类型设计定制化的干扰、噪声抑制方案，提出了周期导向的多层解卷积方法、特征模式分解方法和深度解卷方法等新理论和方法。通过试验研究和工程应用对RV减速器轴承故障诊断方法进行了验证。研究成果将为机器人用RV减速器轴承服役状态监测提供理论和技术支持，具有重要的学术研究意义和工程应用价值。在本项目的资助下，共发表高水平期刊论文13篇，其中第一标注论文11篇；申请和授权国家发明/实用新型专利3项。项目负责人苗永浩，获2022年中国机械工业科学技术发明一等奖(第8)。