An improved LLE algorithm based on iterative shrinkage for machinery fault diagnosis. (January 2016)

• Record Type: Journal Article
• Authors: Liu, Yuanhong; Yu, Zhiwei; Zeng, Ming; Zhang, Yansheng
• Is Part Of: Measurement. Volume 77(2016:Jan.)
• Page Start: 246
• Page End: 256
• Publication Date: 2016-01
• Subjects: Dimension reduction -- Sparse regularization -- Local linear embedding -- Fault diagnosis -- Iterative shrinkage; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2015.09.007
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: ISLLE is proposed with the aid of iterative shrinkage technology and LLE. A surrogate function is introduced to simplify the computational complexity of ISLLE. ISLLE implements feature extraction twice. Based on ISLLE a new fault diagnosis method is developed. Abstract: Local linear embedding (LLE) algorithm is widely utilized to feature extraction for fault diagnosis, but the diagnosis result is sensitive to reconstruction weight W of LLE. To make W more significant and robust, in this paper, ISLLE algorithm is proposed with the aid of iterative shrinkage technology and LLE algorithm. In ISLLE algorithm, a surrogate function is introduced, upon which the high-dimensional optimization problem can be decoupled into a set of one-dimensional equations, then W can be easily 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. Hence, the signals processed by ISLLE are more beneficial to diagnosis. Three real datasets are used to examine the proposed method. The experimental results demonstrate that the proposed method is valid, and the performance of ISLLE outperforms that of the original LLE.