About this document

Predictive Maintenance for UAS RUL Estimation by Clovis Neto is a document available to read on EtoBox.

This document proposes a new data-driven method to estimate remaining useful life (RUL) for a multi-rotor unmanned aircraft system (UAS) using vibration data. It extracts a novel feature called mean peak frequency from vibration signals using short-time Fourier transform. A long short-term memory model is used to forecast future mean peak frequency values and estimate RUL. Testing on three replications achieved RUL estimates of 4, 10, and 10 seconds with root mean square errors of 3.7142, 1.4831, and 1.3455

Author

Clovis Neto

Language

EN