基于多元经验模态分解的电力系统强迫振荡源定位

Forced Oscillation Location in Power Systems Using Multiple Empirical Mode Decomposition

强迫振荡是威胁电力系统安全稳定运行的重要因素之一,准确、高效定位强迫振荡源是抑制强迫振荡的关键,但现有基于广域量测信息的强迫振荡源定位方法实现较为复杂且效率和精度有待提升。为此,该文提出一种基于多元经验模态分解的电力系统强迫振荡源定位方法。该方法首先对发电机的广域量测信息进行多元经验模态分解;然后,计算分解后各本征模函数分量的Teager能量值及能量权重,提取表征强迫振荡模式的关键本征模函数分量;基于提取的本征模函数分量,计算各发电机的耗散能量流,进而根据耗散能量流实现强迫振荡源定位;最后,分别利用WECC179-节点测试系统仿真数据和辽宁电网PMU实测数据对所提方法进行分析验证,结果表明,所提方法可准确定位电力系统强迫振荡源。

Continuous forced oscillation will threaten the secure operation of power systems. Accurate and efficient forced oscillation source location plays a critical role in mitigating forced oscillation, but the existing locating forced oscillation source methods are very complex. To cope with this shortcoming, multivariate empirical mode decomposition(MEMD) was proposed in this paper to locate the forced oscillation source from the measurement responses. The intrinsic mode functions, which represent different oscillation modes, were decomposed from the multichannel measurement responses by using the MEMD. Then, the IMF associated with the forced oscillation mode was separated from the decomposed IMFs by using the Teager energy operator. Using the separated forced oscillation IMF decomposed from the measured frequency, voltage, active power and reactive power of the generator, the dissipation energy flow(DEF) of generator was calculated, and the oscillation source can be located by using the calculated DEF. Furthermore, the performance of the proposed MEMD-based forced oscillation source location was evaluated by the WECC 179-bus test system and Liaoning power grid. The results confirm that the proposed method can accurately and efficiently locate the forced oscillation source from the measurement responses in the bulk power systems.