Interpretation study of quasi-static uni-axial compression analysis of cellular panels with machine learning application

The study aimed to evaluate the energy absorption behavior of cellular sandwich panels made of banyan wood skins and an aluminum honeycomb core under quasi-static loading conditions. The study analyzed two different setups: Type-I panels, which had banyan wood skin plates whose fibers are aligned in plane to the loading axis, and Type-II panels, which had skin plates whose fibers are aligned perpendicular to the loading axis. Both variants reliably aligned the aluminum honeycomb core with its cell axis parallel to the loading direction. An analysis was conducted on the behavior under quasi-static loading circumstances, and the capacities for absorbing energy were measured. The results showed that the energy absorption capabilities were improved during fibers along the cut (Type-I) situations in quasi-static circumstances. Type-I sandwich panels demonstrated exceptional effectiveness in absorbing impact energy, making them especially suitable for applications. Further interpretation of same is developed based on the application of machine learning algorithm. This algorithm considers the wood and aluminum properties and dimension to generate load v/s displacement behavior. The machine learning algorithm also shows the correlation of predicted data found is 99.92% with respect to actual. The algorithm best suits to find the behavioral pattern without conducting experimentation of specified sandwich panels in future application.