基于RDT和ERA算法的低频振荡模态参数辨识

Identification of Low Frequency Oscillation Modal Parameter Based on RDT and ERA Algorithm

针对目前低频振荡主导模式辨识精确度不高以及预警体系不够完善的问题,将航空航天领域用于分析航天器结构振动特性的特征系统实现算法(ERA)与随机减量技术(RDT)相结合,实现基于环境激励下随机响应的低频振荡模态识别。该方法利用RDT从随机响应系统中提取自由振荡信号,进而采用ERA算法对得到的自由振荡信号进行辨识,可准确获取低频振荡频率、阻尼比等参数。通过4机2区域系统仿真辨识,结果表明了该方法的准确性和有效性;并与传统的Prony方法进行了对比,表明了该方法在辨识精度和抗噪性方面有更好的表现。该方法给在多通道低频振荡模态参数辨识提供了一种更加准确有效的手段,能够满足电力系统低频振荡在线辨识的要求,具备很好的应用前景。

In view of the current low-frequency oscillation dominant mode identification accuracy is not high and the early warning system is not perfect,This paper adopts a research method that combined ERA and RDT,which are previously used to analyze the vibration characteristics of spacecraft structuresin the field of aerospace while now to realize low frequency oscillation mode identification based on random response under environmental excitation.This method uses the RDT to extract the free-oscillation signal from the random response system,and then uses the ERA algorithm to identify the free-oscillation signal obtained,and accurately obtains parameters such as low-frequency oscillation frequency and damping ratio.The simulation results of 4-machine and 2-zone system show the accuracy and validity of this method.Compared with the traditional prony method,this method shows better performance in recognition of accuracy and noise immunity.This method provides a more accurate and effective method for multi-channel low-frequency oscillation modal parameter identification,which can meet the requirements of low-frequency oscillation on-line identification of power system and thus has a good application prospect.