As the increasing capacity of 10 kV distribution network caused a rush rise of the short-circuit current

even exceed the max breaking capacity of fault current limiter（FCL）. The normal algorithms to detect the short-circuit fault are always by current RMS

transient value or slope

but these methods are imperfect in detection speed or detection accuracy. This paper presents a real-time wavelet transform algorithm to detect fault quickly in distribution network. By updating eight current sampling data each time and using the method of recursion wavelet transform

we combine the sum of the absolute value of the third wavelet coefficients d3 to the current instantaneous value as a comprehensive criterion to detect short circuit faults. The three-phase short-circuit fault simulation are carried out based on MATLAB/Simulink

it presents a detailed explanation of the process. Finally

this paper has conducted thousands of simulations of the time of short-circuit fault detection

and the results proved that the real-time wavelet transform can detect short circuit faults quickly and effectively

it also has good adaptability.