中文摘要：

基于端口供给能量分解,提出了一种新的电力系统振荡类型识别方法。对端口供给能量进行成分分解,发现弱阻尼振荡和强迫振荡下端口供给能量趋势分量分别按指数和线性规律变化。采用经验模式分解算法识别出主导振荡模式,计算端口供给能量,利用经验模式分解算法提取端口供给能量趋势分量。对变化幅度明显的端口供给能量趋势分量进行线性和指数拟合并得到拟合误差指标,对拟合数据和原始数据进行相关性计算得到相关度指标。然后利用误差指标和相关度指标判别该趋势分量的变化规律,进而利用变化规律的不同识别振荡类型。最后通过IEEE 39节点系统的仿真分析和实际电厂的实例分析证明了该识别方法的有效性。

英文摘要：

A new method of identifying the power system oscillation property using decomposition of energy supply on the port (ESP) is proposed.Component decomposition of ESP is conducted and found to show that the trend component of ESP changes exponentially under weak damp oscillation while linearly under forced oscillation,respectively.ESP is calculated using the component of dominant mode in electrical variables,which are acquired by empirical mode decomposition (EMD).The trend components of ESPs are also obtained by EMD with the obvious trend component of ESPs fitted by exponential and linear curves,respectively.The correlation coefficients between real trend component and fitting trend component are calculated under two fitting types.The change rule of the trend component is determined by fitting error indices and the correlation coefficient.The oscillation type is identified according to the identified change rule.Finally,the simulation analysis of an IEEE 3 9-bus system and the instance analysis of a real power plant has proved the validity of the proposed method. This work is supported by the Foundation for Innovative Research Groups of the National Natural Science Foundation of China(No.51321005)and National Natural Science Foundation of China(No.51177079).