Establishing Statistical Correlation Between Sensor Signature Features and Lubricant Solid Particle Contamination in a Spur Gearbox

Abstract

This paper aims to predict the severity of solid particle contaminants present in the lubricant in a Spur Gearbox using Vibration, Acoustic Emission, and Sound Signature features. Sensor signatures are acquired at various contaminant conditions of lubricant with different speed and load conditions. Statistical Features are extracted in the time domain, and feature ranking is carried out using the analysis of variance approach. Statistical models are developed using the selected features of Sound, Acoustic Emission, and Vibration separately and fusing the features in the feature level. Decision Trees and Support Vector Machine algorithms are used in this study to build statistical models. AE features have a good correlation with lubricant conditions compared to sound and vibration features. The feature-level fusion approach predicts the lubricant conditions with an accuracy of more than 99%. The feature-level fusion models built using dominant features are computationally efficient without compromising the prediction ability.

Correlative Method for Diagnosing Gas-Turbine Tribological Systems

“…Some issues, related to skewed data distribution, were overcome here by removing outliers and logarithmic transformation. Undoubtedly, this dataset and similar oil analysis results can be processed with more advanced statistical or machine learning models such as quantile regression, decision trees [ 63 , 64 ], SVM [ 65 , 66 ], artificial neural networks [ 67 ], or their assemblies [ 68 , 69 ]. There are enough training data, and the computational cost is moderate, but the main challenge is to properly formulate the prediction problem when engine wear and its remaining useful life is not a priori known.…”

Review of advances in tool condition monitoring techniques in the milling process

“…Amplitude, duration, measured area of the rectified signal envelope (MARSE), and counts are the main signal parameters in AE [33]. The highest voltage in a waveform is measured by its amplitude, which is expressed in decibels.…”

Mapping Acoustic Frictional Properties of Self-Lubricating Epoxy-Coated Bearing Steel with Acoustic Emissions during Friction Test

“…Many studies have previously used acoustic signals to monitor the wear of grinding wheels [23][24][25][26], ball-on-flat sliding contact [27], stick-slip [28], bearings [29], gearboxes [30], mill-grinding tools [31], and tools [32]. AE data extraction methods include fast Fourier transforms (FFT) [24], short-time Fourier transforms (STFT) [25], wavelet transform (WT) [26], amplitude [28], AE count [33], spectral kurtosis [29], and root mean square (RMS) [34], which are used to correlate tribological parameters with acoustic signals.…”