PUDT: Plummeting uncertainties in digital twins for aerospace applications using deep learning algorithms

Abstract

Identifying objects in aircraft monitoring systems poses significant challenges due to the presence of extreme loading conditions. Despite the presence of several sensor units, the transmission of precise data to multiple data units is hindered by an increase in time intervals. Therefore, the suggested methodology is specifically developed for the purpose of generating digital replicas for aeronautical applications, wherein an aero transfer function is correlated with the digital twins. Mapping functions are utilized in the monitoring of diverse parameters that are associated with the identification of objects inside data transmission networks, with the aim of minimizing uncertainty. The suggested system model is enhanced by incorporating analytical representations and deep learning methods, resulting in the provision of zero point twin functionalities. The present study investigates the aforementioned integrated procedure through the analysis of four different situations. In these settings, an aero communication tool box is employed to transform the device configuration into simulation outputs. The results obtained from the comparison of these scenarios reveal that the projected model significantly enhances the maintenance period while minimizing data errors.