Research on the Optimization of BP Neural Network Based on Genetic Algorithm and Its Application

There are some flaws for back propagation(BP)neural network, such as lowconvergence and vulnerable to local minimum terminations. After analyzing the training algorithm nature of BP neural network, genetic algo-rithms(GA)is introduced to optimize the right values and thresholds of BP neural network, and then a new GA-BP neural network is presented, which takes full use of the global optimization of GA and local accurate searching of BP. Application simulation results show that GA-BP neural network can converge at optimum points rapidly, and the built GA-BP neural network model has high prediction precision and good generalization performance.