项目名称: 双转子中介轴承振动故障监测预警原理和方法研究
项目编号: No.51305020
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: 冯坤
作者单位: 北京化工大学
项目金额: 26万元
中文摘要: 涡扇发动机双转子中介轴承振动故障导致的重大事故时有发生,迫切需要自主研发故障监测预警系统,双转子中介轴承振动故障监测预警理论和方法研究十分必要。针对涡扇发动机双转子结构和工况决定的复杂传递路径,以及中介轴承与后轴承故障耦合的特点,提出基于理论分析、建模仿真计算、双转子中介轴承实验台实验,研究中介轴承故障机理,揭示故障弱信号复杂路径的传递规律,探索中介轴承故障与后轴承故障的耦合机制。基于状态监测数据,运用稀疏表达(SpR)和子空间伪框架(PFFS) 等先进信号处理手段,以故障机理和传递规律为指导,研究故障微弱信号特征提取原理和方法。进而基于故障耦合机制和特征提取研究结果,运用深度学习神经网络(DLANN)及支持向量机(SVM)等先进模式识别技术,研究中介轴承早期故障识别原理和方法。为研制涡扇发动机中介轴承振动故障在线监测预警系统提供科学依据和技术支撑。
中文关键词: 中介轴承;复杂传递路径;微弱信号特征提取;稀疏表达;支持向量机
英文摘要: Turbofan major accident caused by double-rotor intershaft bearing vibration fault happens regularly. Independently developed fault monitoring and warning system are urgently desired by China. Hence, principle and method study on intershaft bearing vibration fault monitoring and warning is presented in this project. Complex transfer path, coupling with the rear bearing fault are common properties of double-rotor turbofan engine. The study of this project are conducted according to these properties. Based on theoretical analysis, modeling, simulation, numerical calculation and double-rotor bearing rig experiment, intershaft bearing fault mechanism is studied, fault weak signal transfer law of complex path is revealed, and the coupling mechanism of two bearing faults is researched. Based on condition monitoring data and advanced signal processing methods such as sparse representation (SpR) and Pseudo-frames for subspace (PFFS), weak fault signal feature extraction theory is studied under the direction of fault mechanism and transfer law. Furthermore, based on the research results of fault coupling mechanism and feature extraction, fault recognition method is studied through application of advanced pattern recognition techniques such as deep learning artificial neural network (DLANN) and support vector machine (SVM
英文关键词: intershaft bearing;complex transfer path;weak signal feature extraction;Sparse Representations;Support Vector Machine