基于自适应陷波滤波器和混沌振子的配电网故障选线

Fault line selection in distribution network based on adaptive notch filter and chaos oscillator

小电流系统发生单相接地故障时,各馈线的零序电流呈现出复杂的非线性和非平稳性,故障点的信号电流也非常微弱,给故障选线带来困难。但故障后,故障馈线的零序高频分量幅值是所有正常馈线之和,且相位相反。文中提出一种结合自适应陷波滤波器(Adaptive Notch Filter,ANF)和混沌振子的选线方法。ANF可以通过调节本身参数,实时、准确提取各馈线高频信号,而混沌振子对内外同频率信号有较高敏感性,对其进行尺度变化以适应电网高频信号检测。通过ANF提取故障零序电流中五次谐波,将其作为混沌振子的输入,用获得的系统相图作为故障判据。对不同条件下的故障进行了仿真验证,结果证明了算法的准确性。

When a single-phase-to-ground fault occurs in a small current system,the zero-sequence current of each feeder presents complex nonlinearity and non-stationarity,and the signal current at the fault point is very weak,which brings difficulties to fault line selection.However,the zero-sequence high frequency component of fault feeder is larger than that of normal feeder after fault,and the phase is opposite.In this paper,a line selection method combining adaptive notch filter(ANF)and chaotic oscillator is proposed.By adjusting its parameters,ANF can extract high-frequency signals from feeders in real time and accurately.Chaotic oscillator is more sensitive to the same frequency signals inside and outside,and its scale changes to adapt to high-frequency signal detection of power grid.Firstly,the fifth harmonic of fault zero-sequence current is extracted by ANF,which is used as the input of chaotic oscillator.The obtained system diagram is taken as the fault criterion.The faults under different conditions are simulated and verified,and the simulation results show that the algorithm is accurate.