Zero-crossing Prediction of Short-circuit Current Using Recursive Least Square Algorithm

The fast and accurate prediction of the fault current zero-crossing is the key to controlled fault interruption. According to the characteristics of the short-circuit current signal, this paper uses the recursive least square (RLS) algorithm to estimate the fault current parameters. For the truncation error caused by Taylor series expansion, a correction method of time constant is proposed. The restored waveform of the sampled signal is calculated to predict the zero crossing. An ideal signal is set in MATLAB to verify the performance of the algorithm, and the results show that the RLS algorithm gains small calculation error and high waveform fitting degree, which can accurately predict the zero-crossing point of fault current. Finally, the simulation results of the fault recording data further verify the reliability and practicability of the algorithm.