Open AccessThe Method of BP Algorithm for Genetic Simulated Annealing Algorithm in Fault Line Selection
Author(s) -
Yueyang Song,
Huang Hong-quan,
Yanming Chen
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1650/3/032187
Subject(s) - algorithm , simulated annealing , crossover , artificial neural network , adaptability , computer science , genetic algorithm , selection (genetic algorithm) , convergence (economics) , population , population based incremental learning , artificial intelligence , machine learning , ecology , demography , sociology , economics , biology , economic growth
Aiming at the shortcomings of low accuracy and poor applicability of the traditional ground fault line selection method for small current grounding systems, an BP neural network algorithm based on genetic simulated annealing algorithm (GSAA-BP) is used to select the fault line. This algorithm not only avoids the extortionate proportion of the initial weight and threshold of the traditional BP neural network, but also improves the population diversity by changing the crossover and mutation probability of the genetic algorithm. It avoids the result of the algorithm converge to a local optimum. Comparing with other BP neural network algorithms shows that this method in training has higher convergence speed, better complex adaptability and more accurate judgment accuracy.