Impact energy absorption with artificial neural network (ANN)

Artificial Neural Network is a numerical method for solving complex problems. In the neural network, a relationship is established between the input and output of the problem. The ability of the neural network to solve non-linear and complex problems that are solved using statistical information. Different software such as MATLAB, NeuroSolution, and easynn can be used to design the neural network.

In this project, Neurosolution V5.05 software is used to simulate the impact energy absorption of aluminum/epoxy composite in the Charpy test with artificial neural network (ANN). The inputs of the project were the thickness of the layers, the number of layers, the type of adhesive, the condition of the crack tip, the amount of silicon carbide (SiC), the amount of MBS, and the number of tests performed for each sample.

The output of the project is the energy absorbed by the aluminum/epoxy composite during impact. The data related to the problem has been prepared through laboratory testing. A total of 126 tests have been performed using the Charpy method. The MLP method was used to solve the network. Various functions were used in this problem, but in the end, it was concluded that the best one for this problem is the sigmoid function. The number of layers was considered for analysis, the best of which was one layer, and the best learning method for this was determined conjugate gradient problem.


This model was verified with the data in the Nazari paper.

impact energy absorption of aluminum/epoxy composite in the Charpy test with an artificial neural network (ANN)