基于叶尖定时法的叶片耦合振动参数辨识

发布时间：2018-12-16 21:53

【摘要】：为了实时掌握叶片振动信息预防叶片故障发生,常用的非接触振动测量方法是叶尖定时法,叶尖定时法采集的叶片振动信号为欠采样信号,想要了解叶片的振动信息,其关键技术在于振动参数识别算法的建立。目前常用的都是基于单自由度模型的叶尖定时算法,当叶片发生耦合振动时会出现两个甚至多个振动峰,且相邻两个振动峰相隔很近,这种方法无法同时识别出相邻两个峰的振动参数。采用基于两自由度模型的曲线拟合法,同时拟合出叶片耦合振动两个峰的振动参数,并运用基于两自由度模型的叶尖定时算法对叶片耦合振动进行参数辨识。不仅提高了叶片耦合振动参数辨识的精度,而且可同时辨识出相邻两个峰的振动参数。本文的主要研究工作如下:(1)对基于两自由度模型的叶片耦合振动进行了理论推导,研究了叶片间耦合振动对叶片振动参数识别的影响。开发了一套4叶片仿真软件,可仿真同步振动、异步振动、整级叶片振动等不同类型的叶片振动信号,并对各种仿真信号的产生和特点进行了简单的分析。(2)对基于两自由度模型的曲线拟合进行了理论推导,使其可同时拟合叶片耦合振动相邻的两个振动峰。编写基于两自由度模型的自回归和正弦拟合算法,应用于叶片耦合振动参数辨识。并用仿真数据对这两种叶尖定时算法进行了验证,结果表明辨识振动参数与仿真参数设置基本一致。(3)基于Matlab开发了一套叶片耦合振动参数辨识软件。进行了各种情况下的大量仿真验证,结果正确。软件界面友好、编程规范、辅之以丰富的鼠标和键盘控制,实现人性化操作。
[Abstract]:In order to grasp the blade vibration information in real time to prevent the blade fault, the common non-contact vibration measurement method is the blade tip timing method, the blade vibration signal collected by the blade tip timing method is under-sampled signal, so we want to know the blade vibration information. The key technology is the establishment of vibration parameter identification algorithm. At present, two or more vibration peaks appear when the coupling vibration of the blade occurs, and the two adjacent vibration peaks are very close to each other. This method can not identify the vibration parameters of two adjacent peaks at the same time. The method of curve fitting based on two-degree-of-freedom model is used to fit the vibration parameters of two peaks of coupled blade vibration simultaneously, and the blade tip timing algorithm based on two-degree-of-freedom model is used to identify the parameters of coupled vibration of blade. Not only the accuracy of the coupled vibration parameter identification of the blade is improved, but also the vibration parameters of the adjacent two peaks can be identified at the same time. The main work of this paper is as follows: (1) the coupling vibration of blades based on two-degree-of-freedom model is derived theoretically, and the influence of coupling vibration between blades on the identification of blade vibration parameters is studied. A set of 4-blade simulation software is developed, which can simulate different kinds of blade vibration signals, such as synchronous vibration, asynchronous vibration, whole blade vibration, etc. The generation and characteristics of various simulation signals are analyzed simply. (2) the curve fitting based on the two-degree-of-freedom model is derived theoretically so that it can simultaneously fit the two adjacent vibration peaks of the coupled vibration of the blade. The autoregressive and sinusoidal fitting algorithms based on two-degree-of-freedom model are developed and applied to the identification of blade coupling vibration parameters. The simulation data are used to verify the two algorithms, and the results show that the identification parameters are in good agreement with the simulation parameters. (3) based on Matlab, a set of software is developed to identify the coupled vibration parameters of the blades. A large number of simulations are carried out under various conditions, and the results are correct. Friendly software interface, programming specifications, supplemented by rich mouse and keyboard control, to achieve human operation.
【学位授予单位】：南京航空航天大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：V263.6

【相似文献】