DATA PREPROCESSING IN THE CONTEXT OF PHM -CHALLENGES AND SOLUTIONS
Abstract
This study presents a comprehensive framework for addressing data preprocessing challenges in diverse scenarios. The framework covers various aspects, including feature extraction in the time domain, frequency domain, and wavelet domain, as well as feature selection techniques, noise reduction strategies, and data normalization approaches for time series data. Through experimental analysis, we demonstrate the effectiveness of our proposed strategy in enhancing the Signal-to-Noise Ratio value. In particular, the application of the Savitzky-Golay Filter proves successful in reducing signal noise and achieving normalization across all bearings. These findings highlight the potential of our framework in improving data quality and preprocessing procedures for bearing data set which can be generalized for a wide range of applications.
Fichier principal
DATA PREPROCESSING IN THE CONTEXT OF PHM – CHALLENGES AND SOLUTIONS.pdf (973.94 Ko)
Télécharger le fichier
Origin : Files produced by the author(s)