A prognostic approach to improve system reliability for aircraft system

The primary aims of prognostics encompass the timely detection of potential failures, mitigation or elimination of unscheduled maintenance, prediction of the most suitable timing for preventive maintenance replacement, optimization of maintenance cycles and operational readiness, and enhancement of system reliability by improving design and logistical support for existing systems. In order to facilitate the progress of these approaches, currently available datasets provide a unique and reliable compilation of flight-to-failure trajectories linked to small aircraft engines that have been observed in actual flight conditions. Furthermore, the paper offered an improved neural network that utilized the TanH hyperbolic tangent function. This neural network was enhanced later by integrating it with the TanH, linear, and Gaussian functions. Additionally, a random holdback validation approach was employed in the paper. The results suggest that the NN TanH technique, when implemented, has the potential to significantly enhance the reliability of an aircraft component. This is achieved through accurate estimates of the remaining useful life (RUL) and a proactive understanding of the failure system.