高速旋转叶片异常振动非接触在线检测与诊断技术研究

发布时间：2018-06-22 01:30

本文选题：叶片裂纹 + 异常振动　；参考：《国防科学技术大学》2012年硕士论文

【摘要】：高速旋转叶片是燃气轮机中的关键部件，长期工作在高温、高压、大应力等恶劣环境下，且承受离心力、气动力等动载荷作用，容易产生振动。各种异常振动是造成叶片损伤或故障的主要原因，因此在线检测高速旋转叶片的异常振动具有重要意义。为此，本文针对常见的叶片裂纹故障，以简化叶盘为研究对象，开展了基于叶端定时原理的高速旋转叶片振动信号非接触在线检测与裂纹故障诊断方法研究。主要工作包括：（1）为了分析裂纹对叶片振动行为的影响，，利用Euler-Bernoulli梁理论建立了含裂纹叶片的动力学模型，推导了其动频表达式，并建立有限元模型进行了仿真验证。结果表明：裂纹导致叶片动频减小、振幅增大，并且越靠近叶片根部，影响越大；（2）推导了利用3个叶端定时传感器的高速旋转叶片振动位移表达式，并针对其具有欠采样的特点，研究了基于香农采样定理的信号重构方法和基于平滑的振动信号预处理方法，为后续叶片裂纹特征提取和故障诊断提供了数据基础；（3）分别研究了欠采样信号下基于余数定理的叶片振动频率提取方法、重构信号下基于状态空间模型的叶片模态频率提取方法，并在此基础上研究了基于模态局域化的叶片裂纹故障诊断方法，可有效识别不同的裂纹状态；（4）设计了高速旋转叶片实验台，搭建了叶片振动信号非接触在线检测系统，通过人为方式注入叶片裂纹，对所研方法进行了实验验证；结果表明，本文所研方法可在线识别高速旋转叶片的异常振动、检测不同程度的裂纹故障，可有望用于燃气轮机转子叶片状态的在线监测。
[Abstract]:High speed rotating blade is the key component of gas turbine. It works in high temperature, high pressure and high stress for a long time, and it is easy to produce vibration under the action of centrifugal force, aerodynamic force and other dynamic loads. Various abnormal vibration is the main cause of blade damage or fault, so it is very important to detect the abnormal vibration of high-speed rotating blade online. For this reason, aiming at the common blade crack faults and taking the simplified blade disk as the research object, the method of non-contact on-line detection and crack fault diagnosis of high-speed rotating blade vibration signal based on the principle of blade tip timing is studied in this paper. The main works are as follows: (1) in order to analyze the effect of crack on blade vibration behavior, the dynamic model of blade with crack is established by using Euler-Bernoulli beam theory, the dynamic frequency expression is deduced, and the finite element model is established to verify the dynamic behavior. The results show that the crack reduces the dynamic frequency and increases the amplitude of the blade, and the closer the blade is to the root, the greater the effect. (2) the vibration displacement expression of high-speed rotating blade using three tip timing sensors is derived. The signal reconstruction method based on Shannon sampling theorem and the vibration signal preprocessing method based on smoothing are studied in order to provide the data basis for the crack feature extraction and fault diagnosis of the subsequent blade. (3) the extraction method of blade vibration frequency based on remainder theorem in under-sampled signal and the modal frequency extraction method based on state space model in reconstructed signal are studied respectively. On the basis of this, the fault diagnosis method of blade crack based on modal localization is studied, which can effectively identify different crack states. (4) A high-speed rotating vane experiment bench is designed. A non-contact on-line detection system for blade vibration signal is built. The method is experimentally verified by artificial injection of blade cracks. The results show that, The method proposed in this paper can identify the abnormal vibration of high speed rotating blades and detect crack faults of different degrees on line, which is expected to be used in the on-line monitoring of rotor blade condition of gas turbine.
【学位授予单位】：国防科学技术大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH165.3

【参考文献】