Prediction model of rolling force for electrical steel based on finite element method and neural network

The rolling force prediction model for electrical steel in hot rolling was established. A 3D elastic-plastic finite element model of roll and strip was established by MSC. Marc software package. The influence of strip thickness, rolling temperature, strip width, friction coefficient, reduction and rolling speed on rolling force was attained by simulation. The rolling force prediction model for electrical steel in hot rolling was established based on FEA and BP neural network. The model with high accuracy and calculation efficiency can be used to guide the production.