摘要
采用平滑优先方法(SPM)对电力系统受扰轨迹进行预处理,以消除其中的低频非平稳趋势项,便于电力系统的模态参数辨识。介绍了SPM算法,分析了不同平滑参数下SPM的截止频率,结合电力系统机电振荡频率范围,指出平滑参数取值500左右可有效去除待分析信号中的低频趋势项,同时保留感兴趣的模式。以人工合成信号为例,仿真分析表明,平滑参数不仅对信号中低频模式成分的能量有较大影响,同时还影响低频模式成分中阻尼辨识的精度。进一步以4机2区域系统为例,结果表明SPM可有效剔除信号中低频趋势项,提高模态参数的辨识精度。
The SPM(Smoothness Priors Method) is used to preprocess the real disturbed trajectory to eliminate the non-stationary trending components for the modal parameter identification of power system.The SPM algorithm is introduced and the cutoff frequency is analyzed for different smoothness parameters.According to the frequency range of power system electromechanical oscillation,the smoothness parameter of about 500 may effectively eliminate the low-frequency trending components while reserve the useful components of the signals.Simulative analysis for a synthetic signal shows that,the smoothness parameter has greater influence on both the energy and the damping identification accuracy of low-frequency components.With a 4-machine-2-area system as example,the results show that,SPM can effectively eliminate the non-stationary trending component and improve the accuracy of modal parameter identification.
出处
《电力自动化设备》
EI
CSCD
北大核心
2010年第6期63-66,共4页
Electric Power Automation Equipment
关键词
平滑优先方法
特征根
模态参数辨识
主导振荡模式
PRONY分析
smoothness priors method
eigenvalue
modal parameter identification
critical oscillation mode
Prony analysis