| Title |
Psychoacoustic-based Fault Diagnosis for Drone Blade Damage |
| DOI |
https://doi.org/10.5573/ieie.2025.62.2.67 |
| Keywords |
Fault diagnosis; Blade damage; Psychoacoustic; Acoustic sensors |
| Abstract |
In this paper, data pre-processing and model training are conducted to diagnose drone blade damage. To simulate the damaged blade, a length-based quantitative breakage is applied and data is collected accordingly. The acquired data is processed using STFT and various psychoacoustic/traditional acoustic features to use it for model input. Among these indicators, four were identified as particularly effective in representing blade breakage, enhancing model performance compared to using acoustic signals alone. K-fold cross-validation is performed to validate the proposed model and to show our findings. |