Characterizing the time-dependent external force on the cars’ hood door in accident using deep neural networks

Abstract

The car hood door is a very important part of a car, especially in the time of an accident. So, in this report for the first time, the innovative applicable model is presented to simulate the transient dynamics of a car’s hood door under axial mechanical shock loading at the time of the accident. Due to axial mechanical shock excitation, it is very important to improve the stability of the structure in the axial direction. So, the composite structure is reinforced by graphene nanoplatelets in the axial direction. For modeling the current structure, the differential conditions are translated into Laplace space in order to determine how the system will react as a function of time. At this stage, Laplace space is employed to infer a temporal perception of the system's reaction using Abate and Dubner's modified message of strategy. To verify the results, the current results are compared to open-source results from the literature and deep neural networks (DNN). To anticipate the system's vibrational behavior, DNN incorporates a supervised neural network based on physical data. In this situation, data-driven research and solutions are used to determine natural frequencies. The results highlight how important GPL characteristics are to transient and forced vibrations of the composite system. The findings of the present research may serve as a reference point and as helpful suggestions for the next structural design methods with improved qualities.