An Intelligent Prediction Strategy of the Maximum Thinning Rate of Cylindrical Part with Flange during Sheet Metal Drawing Process

Abstract

In this study, an intelligent strategy of predicting the maximum thinning rate is proposed. Prediction strategy is based on Back Propagation Neural Network (BPNN). Because Blank Holder Force (BHF) can effectively reflect the thinning rate of cylindrical part with flange, the nodes of which are served as the inputs of BPNN in this research. The prediction objective is to obtain corresponding the maximum thinning rate of cylindrical part, which is served as the output of BPNN. The inputs data are derived from designed several critical types of Variable Blank Holder Force (VBHF). In addition, the output data is originated from corresponding simulation results of cylindrical part drawing. The mathematical expressions of prediction process have also been given in this paper. The advantage of this novel prediction strategy is that it can rapidly and accurately obtain the thickness reduction rate of cylindrical part with flange. Eventually, the validation of the created prediction method is proved by corresponding simulations and experiments.