Prediction Model for Impact Energy of Hot Rolled Plate Based on Improved RBF

In order to predict the impact energy of Q345B hot rolled plate,a RBF neural network was used to predict relation between impact energy and the factors of composition,process parameters and strength of the plate.The genetic algorithm was used to optimize the weights and structure of RBF neural network aiming at its shortcomings.The prediction model for impact energy of hot rolled plate based on improved RBF was developed.The experimental result shows that the neural network algorithm optimized by this improved genetic algorithm is effective and feasible.