Enabling Health Monitoring Approach Based on Vibration Data for Accurate Prognostics

Abstract

Performances of data-driven prognostics approaches are closely dependent on form, and trend of extracted features. Indeed, features that clearly reflect the machine degradation, should lead to accurate prognostics, which is the global objective of the paper. This paper contributes a new approach for features extraction / selection: the extraction is based on trigonometric functions and cumulative transformation, and the selection is performed by evaluating feature fitness using monotonicity and trendability characteristics. The proposition is applied to time-frequency analysis of non-stationary signals using Discrete Wavelet Transform. The main idea is to map raw vibration data into monotonic features with early trends, which can be easily predicted. To show that, selected features are used to build a model with a data-driven approach namely, the Summation Wavelet-Extreme Learning Machine, that enables a good balance between model accuracy and complexity. For validation and generalization purpose, vibration data from two real applications of Prognostics and Health Management challenges are used: 1) cutting tools from Computer Numerical Control (CNC) machine (2010), and 2) bearings from platform PRONOSTIA (2012). Performances of the proposed approach are thoroughly compared with the classical approach by performing: feature fitness analysis, cutting tool wear "estimation" and bearings "long-term predictions" tasks, which validates the proposition.

Remaining useful life prediction of the ball screw system based on weighted Mahalanobis distance and an exponential model

“…In this section, the original features are obtained with time-domain techniques, time-frequency-domain techniques, and trigonometric functions [35].…”

“…In addition to the above features, two features are also extracted with trigonometric functions. Specifically, the raw vibration signals are first transformed to different scales using trigonometric functions, and then the standard deviations (SDs) of the scaled sequences are calculated as features [35]. In this research, two trigonometric functions, the inverse hyperbolic sine and the inverse tangent, are selected to process the vibration data, and the two corresponding features are obtained using the following two equations:…”

A Novel Image Feature for the Remaining Useful Lifetime Prediction of Bearings Based on Continuous Wavelet Transform and Convolutional Neural Network

“…Time domain features are suitable for stationary signals. However, they may be sensitive to changes in the signal and could inherit nonlinearity [12]. Frequency domain features can describe additional information that cannot be observed in the time domain because of their superior ability to identify and isolate frequency components [13].…”

SVM hyperparameters tuning for recursive multi-step-ahead prediction

“…Prognostics and health management (PHM) is becoming an important topic of interest for industrial applications [20,24,25,40,45,48]. Prognostics includes the prediction of the health condition of the equipment of interest in the future [19,33,35,49].…”