Finite Element Modeling and Prediction of Thickness Strains of Deep Drawing using an ANN for ASS304

摘要

In this paper, an Artificial Neural Network (ANN) is used to predict the thickness strains along the deep drawn cup. LS-Dyna Simulation results at various parameters are used to train the network, and then the model was applied to new data to predict the thickness strains along the formed cup. These predicted values are compared with the experimental values. A comparison is made between the experimental, simulated and ANN predicted values. ANN predictions are fairly in accordance with experimentally found thickness strains. Application of ANN helps for a faster approach to the required optimum final product, avoiding the time consuming experimentation of the process. This approach helps in narrowing down the wide spectrum of varied range of parameters, roughly estimating the optimum values and also the numbers of simulations are reduced, hence reducing the experimentation.