Abstract
The locally linear embedding (LLE) algorithm can effectively extract low-dimensional feature embedded in the high-dimensional nonlinear data. In this paper, we propose a novel approach of fault diagnosis based on LLE algorithm, introducing LLE into equipment fault diagnosis field and solving fault pattern classification problem. A nonlinear dimensionality reduction algorithm based on LLE is utilized to learn original fault signal directly and extract intrinsic manifold feature in data set. The proposed approach can greatly hold the global geometry structure information embedded in the signal, and availably overcome the flaw of traditional pattern recognition methods which only obtain datapsilas local linear structure, obviously improve classification performance of fault recognition. Experiments with simulation and engineering instance illustrate the feasibility and effectiveness of the new approach.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have