Fault detection in non-Gaussian vibration systems using dynamic statistical-based approaches

Abstract

This article develops and contrasts two different statistical-based techniques for monitoring mechanical systems that produce stochastic, non-Gaussian, and correlated vibration signals. Existing work in this area relies on the assumption that the recorded signals follow a multinormal distribution and/or the data model is static, i.e. the signals are assumed to possess no serial correlation. The developed approaches rely on (i) recent work on independent component analysis and support vector data description that is applied to a dynamic data structure and (ii) the incorporation of the statistical local approach into a dynamic data representation. The analysis of experimental data from a gearbox system confirms (i) significant auto- and cross-correlation within and among these signals and (ii) that they cannot be assumed to follow Gaussian distributions. The application of both approaches showed that they are more sensitive to incipient faults than conventional multivariate statistical methods.