适用于电力系统非平稳功率振荡信号特征提取的自适应迭代滤波算法研究

Extracting Oscillation Modes From Nonstationary Signals for Power System Using Adaptive Local Iterative Filter

规模化新能源并网及大容量电力电子设备的应用使得电力系统功率振荡表征出很强非平稳特性。该文提出了一种适用于电力系统非平稳功率振荡信号特征提取的自适应迭代滤波算法。自适应迭代滤波算法利用Fokker-Planck方程构建滤波函数,经滤波筛选获取具有平稳特征的本征模态分量,算法具有坚实的数学基础,能够有效避免经验模态分解算法存在的模态混叠问题。首先利用自适应迭代滤波算法分解得到振荡分量,对各分量进行Hilbert变换,实现振荡特征参数的提取。测试信号、仿真信号及实测数据分析结果证明了该文方法的有效性,与经验模态分解的对比结果充分验证了该方法在电力系统非平稳功率振荡特征提取中的强适应性。

With the widespread use of renewable energy and high-capacity power electronic equipment, the non-stationary feature of active power oscillation is strengthened. An adaptive iterative filter decomposition （AILFD） was proposed to identify the parameters of non-stationary oscillating signal in power systems. It utilized Fokker-Planck function to construct filter function and sifted stationary intrinsic mode function by filtering. This method had a solid foundation of mathematics, which can avoid mode-fixing issue in empirical mode decomposition （EMD） effectively. Firstly, AILFD was employed to extract the oscillating components, and then Hilbert transform was be utilized to identify the oscillating characteristic parameters from the intrinsic mode function （IMF）. The availability of the proposed method is confirmed by the results of test signal, simulation signal and measured data. The contrasted results between proposed method and EMD illustrate the adaptability of AILFD for identifying the parameters of non-stationary oscillating signal in power system.