The use of sandwich structures has spread to a wide range of fields today. As a result of their thin top and low stiffness, sandwich beams are very susceptible to local loads. One practical solution to increase the sandwich structure's indentation resistance is to add bending strength to the top. The structural properties of carbon nanotubes make them well-suited for use in polymer matrixes; their addition improves the attributes of polymers. Experimental and theoretical investigations of the sandwich structure's indentation behavior are presented in this article. As well, simulations of sandwich structures are conducted using Abaqus software. A tensile test of composite samples and Mori Tanaka's theory were used to determine the elastic modulus of epoxy resin reinforced with carbon nanotubes at different mass percentages. The elastic modulus of composites can be improved by adding carbon nanotubes up to 0.3 % by mass while the mechanical properties are reduced when more than 0.3 % of carbon nanotubes were added. Finally, the analytical solution and Abaqus modeling results for the indentation of the sandwich structure were compared with the results obtained from the experimental tests. A machine learning model is proposed for prediction of sandwich beams deflection. The results of the predictions were phenomenal, with MAE of less than 2 %. The computational costs have been compared with the experimental results, and the machine learning method is very efficient.
Read full abstract