Research on Prediction Model of Bending Force Based on BP Neural Network with LM Algorithm

Hydraulic bending roller is a most basic and important method for shape control of strip. The rolled shape quality is decided by the setting value of bending farce in great part. This paper chooses five-stand hot tandem rolling mill in 1810 product line of Tangshan Iron and Steel Company as background, and deals primarily with the study of the bending force prediction model of the rolling unit. To counter the imperfection of traditional prediction model and according to feature of hot strip mill, the various factors influencing bending farce are analyzed, and a bending farce prediction model based on BP neural network with LM algorithm is set up. The training and testing simulation for the neural network is done by using the actual production data of hot rolled steel SS400. By means of the analysis toward simulation results, it is shown that the neural network prediction model for bending force has not only a fast convergence speed, but also a high prediction accuracy to meet actual production request. The research provides a direction and foundation for the setting of practical bending force of 1810 hot rolling line.