摘要
针对不稳定信号PRONY辨识的噪声敏感和辨识精度低的局陷,提出一种改进的低频振荡模式辨识方法,实现了在复杂噪声干扰情况下的低频振荡模式的准确辨识。该方法基于EEMD算法设计改进了EEMD滤波器,可以有效滤除噪声,保留更多的信号模态特征。对滤波后的信号采用基于总体最小二乘法-旋转不变技术的信号参数估计(TLS-ESPRIT)算法进行参数辨识,从而获取低频振荡各个模式信息。通过算例仿真验证了该方法的有效性。
Aiming at the limits of Prony method that is sensitive to noise and of low accuracy,an improved method is proposed for identifying low frequency oscillation modes in this paper,which realizes accurate identification of oscillation modes with mass of noise.The method is based on EEMD algorithm to design improved EEMD filter,which can effectively suppress the noise and remain the main characteristic components of signal.TLS-ESPRIT algorithm is applied to detect the noise signal and obtain each of the low frequency oscillation modes.The result of the simulation proves that this method is effective and useful.
出处
《陕西电力》
2012年第7期23-27,共5页
Shanxi Electric Power
基金
国家自然科学基金项目(51077103)
