Lock-out Algorithm of Active-Intervention-Type Arc Suppression Device Based on ELM

The active-intervention-type arc suppression device can effectively prevent arc re-ignition and arc-overvoltage at the fault point in the distribution network. However, under some unfavorable working conditions, it is difficult for the device to extinguish the arc, so the corresponding equivalent circuit is established to derive the expression of the fault current. The expression shows that when low resistance ground fault occurs at the end of a long transmission line with a heavy load, the fault current may increase, so the device needs to be locked in these working conditions. The impact factors of fault current are used as the input characteristics to train the extreme learning machine (ELM) neural network, and then the performance of the neural network is verified by the test set. Finally, the simulation results show that the proposed device locking algorithm based on ELM has a high accuracy, which makes the device locked under unfavorable working conditions.