面向燃机多工况宽频域工作模态参数识别的随机子空间方法  

Stochastic subspace method for identification of operating modal parameters of gas turbine under multi-working condition in wide frequency domain

作　　者：左彦飞 庞陈意 江志农[2] 冯坤[3] ZUO Yanfei;PANG Chenyi;JIANG Zhinong;FENG Kun(MOE Key Lab of Engine Health Monitoring-Control and Networking,Beijing University of Chemical Technology,Beijing 100029,China;Beijing Municipal Key Lab of High-end Mechanical Equipment Health Monitoring and Self-Recovery,Beijing University of Chemical Technology,Beijing 100029,China;China Aero Engine Vibration Health Monitoring-Control Joint Lab,A VIC Shenyang Engine Design Institute-Beijing University of Chemical Technology,Beijing 100029,China)

机构地区：[1]北京化工大学发动机健康监控及网络化教育部重点实验室,北京100029 [2]北京化工大学高端机械装备健康监控及自愈化北京市重点实验室,北京100029 [3]中国航发动力所-北京化工大学航空发动机振动健康监控联合实验室,北京100029

出　　处：《振动与冲击》2023年第7期225-236,311,共13页Journal of Vibration and Shock

基　　金：国家自然科学基金青年科学基金(51905025)。

摘　　要：面向燃气轮机工作状态下多工况、宽频域模态参数识别需求,在对典型燃机整机振动模态分析的基础上,虑及测试数据类型、测点位置与方向的选取及不同运行工况影响,提出一种针对燃机整机工作模态参数识别的随机子空间方法。基于实测振动数据,对典型燃机工作模态参数进行了自动划分识别。结果表明:通过分别控制行块数,充分挖掘位移、速度、加速度三种类型数据包含的不同频段模态参数信息并对识别结果择优合并,能够较好识别出数十倍于燃机工频的宽频域模态;合理选取测点位置和方向,能够得到关心频段内局部或整体模态;利用多转速运行工况数据,能够区分出受转速影响的整体模态。实现了对燃机整机振动宽频域范围内整体和局部模态、机匣和转子模态的识别,可为在此基础上的动力特性分析、整机动力学模型修正、结构振动状态评估、振动故障特征提取等提供支撑。Here,to meet requirements of gas turbine modal parametric identification under multi-working condition in wide frequency domain,a stochastic subspace method for global operating modal parametric identification of gas turbine was proposed based on vibration modal analysis of typical gas turbine considering selection of test data type,measured point location and direction as well as effects of different working conditions.Based on measured vibration data,operating modal parameters of typical gas turbine are automatically divided and identified.The results showed that by controlling number of line blocks respectively,fully mining modal parametric information in different frequency bands contained in three types data of displacement,velocity and acceleration,and combining identification results,modes in wide frequency domain with width being tens of times of gas turbine operation frequency can be better recognized;local or global modes in concerned frequency band can be obtained by reasonably selecting position and direction of measured points;using data under multi-rotating speed operation conditions,global modes affected by rotating speed can be distinguished;identification of global and local modes of overall vibration of gas turbine in wide frequency domain as well as casing and rotor modes is realized to provide supports for dynamic characteristic analysis,dynamic model modification of overall machine,structural vibration state evaluation,vibration fault feature extraction,and others.

关 键 词：燃气轮机 模态参数识别 随机子空间识别方法 宽频域 多工况 

分 类 号：TK47[动力工程及工程热物理—动力机械及工程]