基于奇异值分解理论拓展的转子滚动轴承系统故障特征提取技术研究

项目名称： 基于奇异值分解理论拓展的转子滚动轴承系统故障特征提取技术研究

项目编号： No.51305392

项目类型： 青年科学基金项目

立项/批准年度： 2014

项目学科： 机械、仪表工业

项目作者： 从飞云

作者单位： 浙江大学

项目金额： 25万元

中文摘要： 本研究针对旋转机械中关键组件之一转子滚动轴承系统的运行过程，深入分析转子滚动轴承系统内在耦合机制，揭示耦合条件下的故障振动信号响应特征，并以此类故障特征为基础开展了基于滑移向量序列奇异值分解的故障特征提取技术研究。主要研究内容有：以转子滚动轴承系统故障振动响应为背景，研究转子非线性影响下的滚动轴承故障振动信号建模技术；基于滑移向量序列奇异值分解的故障特征提取技术研究，具体包括滑移向量序列架构机理及其特征分析，以滑移参数和特征序列计算为过程的故障特征提取方法；故障特征匹配和时域特征局部定位技术研究；滑移向量逆向重构技术研究。本研究工作源于学科技术前沿，同时紧密结合工程应用实际的迫切需求，对于提高旋转机械常用部件转子滚动轴承系统的故障特征识别能力具有重要的理论价值，可有效提高旋转机械设备的运行安全可靠性，满足机械、能源、国防等领域对设备安全保障的要求，具有重要的理论意义和应用价值。

中文关键词： 故障诊断；特征提取；信号处理；奇异值分解；滑移向量

英文摘要： The research is focused on the operation process of rotor rolling bearing system which is one of the key components in rotating machinery. The coupling mechanism and fault vibration signal response characteristics of rotor rolling bearing system are analyzed. Fault feature extraction technology based on Singular Value Decomposition (SVD) of slip vector series will be developed in this research. The detailed contents are listed as follow: The rolling bearing fault vibration signal modeling techniques which is influenced by the rotor nonlinear factor. The fault feature extraction techniques based on the slip vector series is studied which includes: Mechanism and characteristics analysis of slip vector series architecture; Fault feature extraction method based on the slip parameters and feature series calculating. Fault feature matching and time domain local characteristic tracking technique research; research of reverse reconstructs of slip vector technique. This research work is proposed according to the analysis of newest research findings and tightly combined with the urgent needs of engineering application. The proposed method can enhance the fault feature identification ability of rotor rolling bearing system. The operation safety and reliability of rotating machinery can be well improved by the project. The

英文关键词： fault diagnosis；feature extraction；signal processing；singular value decomposition；slip vector