Artificial Neural Network-based Load Prediction using Stress-Stiffening Effect

In structural engineering, the determination of external loading is essential under actual working conditions. But experimentally determination of external force is a difficult task. Mostly, to find external forces transducers are used, but mounting it on the structure during assembly or working is not an easy task. In this work, some other alternative to find external force is used with the help of stress stiffening effect. Transverse natural frequency data of a stainless-steel wire is recorded for different axial loading conditions. Then acquired data was given as input to ANN for training, based on feed-forward backpropagation method. An unknown load was calculated using the developed neural network with an estimated error of 9% as compared to the actual values.