A Zero-Shot Intelligent Fault Diagnosis System Based on EEMD

Abstract

With the application of artificial intelligence in modern industry, the demand for intelligent fault diagnosis systems is increasing, which has attracted much attention of researchers. There is a popular methodology to use the knowledge or rules defined by experts for fault diagnosis. However, the expert defined information that needs to be manually annotated for faults by human with expertise knowledge costs too expensive in real industrial scenarios and is not always practicable to unknown fault category. In order to realize the zero-shot intelligent fault diagnosis and extend to a more general case, we propose an intelligent fault diagnosis system in this paper, using the statistical attributes automatically extracted from the vibration signal-self. To obtain the statistical attributes, the signals are decomposed by ensemble empirical mode decomposition (EEMD) to get their detailed time and frequency characteristics. And the statistical description of the time and frequency domains is taken as the attributes to relate the unseen categories with the seen ones. With the attributes, Gaussian models of the seen categories are established. And those of the unseen category are estimated by the attribute strength relation between the seen and unseen categories. The unseen category is finally predicted by maximum a posterior probability (MAP). The experiments on fault dataset show that the proposed system can obtain well recognition performance compared with the classic methods. And the performance can be further improved through attribute selection.

An intelligent compound fault diagnosis method using generalized zero-shot model of bearing

“…Binary classifiers [35] This method employs binary-valued artificial semantic definition for GZSL and corrects semantic descriptions of seen faults by adjusting the error rate of predicted semantic attributes, achieving a more accurate representation. EEMD [36] Integrated Empirical Mode Decomposition is employed to decompose the signal, revealing detailed time-frequency characteristics. Statistical descriptions of the time and frequency domains are utilized as attributes to establish connections between unseen and seen categories.…”

Non-parametric Ensemble Empirical Mode Decomposition for extracting weak features to identify bearing defects

“…Although, EEMD was developed a decade ago, however, it has still been applying for decomposing the signal. Lu et al [26] proposed the zero-shot intelligent fault diagnosis scheme and extracted statistical attributes of the decomposed signals from EEMD and finally obtained time and frequency characteristics. With the attributes of known conditions, the authors constructed the well-known Gaussian model for the purpose of classifying the attributes of unknown conditions.…”