Series fault arc simulation method based on neural network black box model

• Main Authors: WANG ZHIYONG, XUAN FUQIANG, LIU YANLI, XU XIANNENG, LI YING, LI GUOHUA, GUO FENGYI
• Format: Patent
• Language: Chinese; English
• Published: 22.05.2018
• Subjects: MEASURING; MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES; PHYSICS; TESTING

The invention belongs to the field of fault arc simulation and particularly relates to a series fault arc simulation method based on a neural network black box model. The method comprises the following steps: 1) a series fault arc experiment system is used for a series fault arc experiment; 2) Mayr-Schwarz fault arc mathematical model parameters taum, alpha, PS and beta under different experimentconditions are calculated; 3) the optimal value g0u of initial arc conductivity is calculated; 4) a neural network black box model for predicting the parameters of the Mayr-Schwarz fault arc mathematical model and the optimal value of the initial arc conductivity is built; and 5) a series fault arc simulation model is built, and fault arcs can be simulated and analyzed. In comparison with the prior art, the method can predict the series fault arc mathematical model parameters under different circuit conditions, the series fault arc system simulation model can be further built, and a new idea is opened for unable featu