An improved LLE algorithm based on iterative shrinkage for machinery fault diagnosis

CATEGORY: FAULT DIAGNOSIS
An improved LLE algorithm based on iterative shrinkage for machinery fault diagnosis

Type	Journal Article
Author	Yuanhong Liu
Author	Zhiwei Yu
URL	http://www.sciencedirect.com/science/article/pii/S0263224115004728
Volume	77
Pages	246-256
Publication	Measurement
Date	January 2016
Abstract	Local linear embedding (LLE) algorithm is typically used to feature extraction for fault diagnosis. However, diagnosis results are sensitive to reconstruction weight W of LLE. To make W more significant and robust, authors present the ISLLE algorithm with the aid of iterative shrinkage technology and LLE algorithm. In the ISLLE algorithm, a surrogate function is introduced, upon which the high-dimensional optimization problem can be decoupled into a set of one-dimensional equations. W can then be computed by iterative shrinkage method. In each iteration, the small and negative weight coefficients are eliminated, while the large ones are shrunk, which can be regarded as feature extraction and noise reduction. Accordingly, the signals processed by ISLLE are more beneficial to diagnosis.