Automatic segmentation and prognostic method of a turbofan engine using manifold learning and spectral clustering algorithms

Abstract

In most of the previous fault diagnostic literatures, the fault modes and states are pre-determined (i.e. the model structure (topology) is a priori known). However, in practical situation, the monitoring data, especially for the entire life-cycle data, nothing is known about the nature and the origin of the degradation (i.e. the model structure is unknown). Moreover, there is no consensus, how to determine the optimal model structure. In this condition, the different model structures may lead to different fault diagnosis/prognosis results. To address the optimal structure-selection problem, this article presents an automatic segmentation method based on Laplacian eigenmaps manifold learning and adaptive spectral clustering algorithms. Given an entire lifetime data of turbofan engine, we attempt to automatically segment the data into a sequence of contiguous regions corresponding to the degradation states. Furthermore, intrinsic dimensionality estimation, nonlinear dimension reduction, and the optimal number of degradation state estimation have been implemented. Automatic segmentation is applied for degradation state segmentation of non-label life-cycle data, and the output can be considered as the available information for developing fault diagnosis/prognosis. The experimental verification results indicate that the proposed automatic segmentation method is highly efficient and feasible for automatically determining the optimal model structure.

Practical health indicator construction methodology for bearing ensemble remaining useful life prediction with ISOMAP-DE and ELM-WPHM

“…The intrinsic dimensions of sensitive degenerate features are estimated by our previously proposed intrinsic dimensionality estimation method [22]. The dimensions of the selected sensitive features are reduced to d-dimensions using the ISOMAP algorithm, and the optimal weights of each reduced feature are calculated by DE.…”

Achieving a representation of asset data conducive to machine learning driven predictive maintenance

“…Also, signal processing methods are often applied for HI construction [110]. Hu et al [108] constructed a HI to represent a bearing system's conditions using the root mean square (RMS) of representative data [112,113]. In Chapter 9, this work provides a data-driven HI construction method that models the system's decay in reference to the known normal state using relative entropy.…”