Research on start criterion of automatic reclosing of low voltage circuit breaker based on wavelet packet and neural network

After the 10kV distribution network trips due to fault, the 0.4kV low-voltage circuit breakers in each distribution room can not avoid the reclosing time of 10kV distribution network, resulting in the loss of voltage fault or under voltage fault, resulting in the collective trip of low-voltage circuit breakers. Because the low-voltage circuit breaker does not have the function of automatic reclosing, manual on-site closing greatly reduces the reliability of low-voltage distribution network power supply. In order to realize the function of automatic reclosing of low-voltage circuit breaker and improve the reliability of power supply in low-voltage distribution network, a new criterion of automatic reclosing of low-voltage circuit breaker based on wavelet packet and neural network is proposed in this paper. Firstly, all kinds of fault current signals of low voltage distribution network are decomposed by wavelet packet, and the energy eigenvalues of each frequency band are used as the input of BP neural network. Then determine whether the faults of the distribution network occur in the high voltage side or the low voltage side, and determine their fault types, and together with the input voltage signal of the circuit breaker to form the low voltage circuit breaker automatic reclosing start decision criteria. Finally, through modeling and simulation verification, the results show that the method is simple in principle and easy to implement, and can be used as the automatic reclosing criterion of low-voltage circuit breaker to improve the reliability of low-voltage distribution network power supply.