基于叶尖定时原理采集的旋转机械叶片振动信号是离散的欠采样信号,利用插值法对其重建的时域信号能反映叶片振动峰值等参数特征,为故障诊断提供依据。通过分析简谐振动的数学模型和信号采样及重构定理,提出在匀转速下利用若干均匀安装在机匣上的叶尖定时传感器采集叶片振动信息进行振动位移信号重建的新方法。利用高阶B样条函数构建的插值核函数具有快速收敛的特点,在减小信号截断误差的同时降低了算法复杂度,利于工程应用。介绍信号重建误差的来源,其测量精度对转速稳定性有较高要求,通过计算机仿真和模拟振动平台气激试验分析该方法性能,验证其在实践中的有效性。
The non-contact measurement vibration signals obtained by blade tip-timing sensor mounted on the machinery case are discrete undersampling signals.A method for the reconstruction of rotating blade vibration signal with interpolation technique at even speed is presented.The reconstructive signal is helpful for the identification of the blades vibration amplitude which is one of the features for rotating assemblies fault diagnosis.With the analysis of signal's sampling and reconstruction model,the reconstruction is based on a finite number of samples by timelimited kernels which are constructed with the involvement of B-splines.The reconstruction has lower truncation error and less calculation as a result of the rapid constringency of higher-order B-splines.The simulation shows that the precision of signal reconstruction is highly dependent on the stability of sampling rate,which is related with the rotation rate.Gas excitation experimentation on the blade vibration test bed is introduced to test and verify the performance of this method in practice.