【摘要】 监测叶轮旋转机械叶片异常振动，能有效保障叶片的工作安全和预防事故的发生.当前，主流测振采用的是叶尖定时技术，而叶尖定时技术属于一种欠采样技术，要配以相应的参数辨识算法获得所需的振动参数.就现阶段而言，叶尖定时各参数辨识算法均是有键相的，需要安装转速传感器.转速传感器受限于自身安装空间和环境，存在诸多的隐患.无键相测振系统去除了转速传感器，可以避免此类情况发生.基于叶尖定时技术和有键相双参数辨识方法，提出无键相下双参数方法辨识出叶片同步振动的振幅和激励倍频等参数.以实验室叶盘实验台为原型建立失谐叶盘系统模型并进行仿真，验证该方法可行.在有键相和无键相下对参数辨识结果进行比较，检验出无键相双参数辨识方法精度符合要求.更多还原

【Abstract】 Monitoring the abnormal vibration of the blade of impeller rotating machinery can effectively guarantee the work safety of the blade and prevent the occurrence of accidents.At present, the main vibration measurement is the blade tip timing technology.Blade tip-timing is an under-sampling technique.The required vibration parameters should be obtained with the corresponding parameter identification algorithm.At present, once-per revolution sensor should be installed for each parameter identification algorithm of blade tip timing.The speed sensor is limited by its own installation space and environment, there are many hidden dangers.This method in this articl requires no once-per revolution sensor.Based on the blade tip-timing and the traditional two-parameter identification method, this articl proposes a two-parameter method without once-per revolution sensor to identify the parameters of blade synchronous vibration, such as amplitude and excitation frequency multiplier.The model of the detuned blade-disc system was established based on the laboratory blade-disc test bench and the simulation was carried out to verify the feasibility of the proposed method.The results of parameter identification with once per revolution sensor and without the sensor are compared.It is verified that the accuracy of the two-parameter identification method without once per revolution sensor meets the requirements.更多还原