Neural network approach for prediction of deflection of clamped beams struck by a mass

The purpose of this work is to establish an empirical relationship and neural network for the prediction of deflection of clamped metallic beams struck by mass and causing large inelastic deformations. A multivariable power series was selected as the form of the regression model to develop the empirical relationship. Material properties and geometry of both the striker and beam were selected as the independent variables of this model to predict the deflection in beam. Good agreement between the experimental results and the prediction of maximum deflections for various impact energies has been obtained. The data used in the development of statistical model was reanalyzed for the prediction of maximum deflection by employing the technique of neural networks with a view towards seeing if better predictions are possible. The neural network models resulted in very low errors and high correlation coefficients as compared to the regression based models.

► An empirical model and neural networks are developed for the prediction of deflection of clamped beams struck by mass.
► Material properties and geometry of both the striker and beam are used as the independent variables.
► Good agreement is obtained with experimental results.
► Neural network models resulted in very low errors and high correlation coefficients.