Abstract
Degradation modeling is critical for system prognostics and evolution mechanism analysis. This paper introduces a deep learning-based data fusion method for constructing a health index (HI) through the fusion of multiple sensor signals. The proposed method leverages adversarial networks and an RMSprop-based sampling algorithm to model nonlinear relations and improve the stability of the algorithm. Simulation studies and a case study on aircraft engine degradation demonstrate significant improvements in remaining useful life (RUL) prediction over existing methods.
Juan Du
Assistant Professor
My research interests include knowledge-infused data science for quality improvement, industrial data analytics and machine learning, and system informatics and control for manufacturing applications.