A New Data-driven Roll Force and Roll Torque Model Based on FEM and Hybrid PSO-ELM for Hot Strip Rolling

Abstract

In the control system of hot strip rolling, the prediction accuracy of roll force and roll torque model play a key role in the calculation of rolling schedule, which directly affects the thickness accuracy of strip and the stability of strip piercing in finishing rolling area. The commonly used methods for calculating roll force and roll torque are engineering method, energy method, finite element method (FEM), boundary element method and artificial intelligence method. Karman was the first researcher to use engineering method to study the rolling force. He put forward the Karman equation named by his name which was based on the assumed condition of the rolling process, mechanical equilibrium equation, the plastic equation, the contact arc equation and friction law. Orowan1) assumed that sliding occurs when the friction force between the roll and the strip is less than the shear strength of the strip material, and adhesion occurs when the friction force is equal to the shear strength of the material. On this basis, the differential equation of the rolling unit pressure distribution was established combining the plate compression study of L.Prandtl. Sims2) developed analytical expressions of pressure distribution, roll force and roll torque by avoiding most of the numerical integration in Orowan’s theory. Ford and Alexander3) proposed a widely used rolling force model based on the modified Orowan model. The principle of the energy method is to establish the A New Data-driven Roll Force and Roll Torque Model Based on FEM and Hybrid PSO-ELM for Hot Strip Rolling