Parameter identification by neural network for intelligent deep drawing of axisymmetric workpieces

Intelligent deep drawing for axisymmetric workpieces is an important research field of intelligent sheet metal forming, and real-time identification of parameters is a key technology for intelligent deep drawing. This paper presents a feed-forward neural network model based on the LM algorithm (put forward by Levenberg and Marquardt), which is established to realize real-time identification of material properties and friction coefficient for deep drawing of an axisymmetric workpiece. Compared with the previous BP model (neural network based on back propagation algorithm) and GA-ENN (evolutionary neural network based on genetic algorithm) model, the error goal of parameter identification by the LM model is stepped downward to a new level. Therefore, accurate parameter identification, which provides preconditions as well as assurance for accurate prediction and control, lays the basis for intelligent deep drawing of sheet metal.