Degradation Change-Point Detection Based on BEAST Time Series Decomposition  
Author Yu Zhou

 
Co-Author(s) Shen-yan Liu

 
Abstract The variety of operational conditions among comparable systems in a fleet leads to the creation of numerous samples (having multiple degradation paths) and information regarding system performance (featuring multiple state variables) within the fleet. A common technique for modelling degradation variation patterns in such fleets is functional principal component analysis, albeit often resulting in a loss of information on mutations related to the degradation of the system. This paper proposes a method to mine degradation variation patterns through BEAST time series decomposition integrated functional clustering. The assumption is that functional characteristics evolve in the degradation paths of repairable systems, prompting the utilization of functional data analysis methods for clustering the corresponding degradation variation patterns. To analyze the impact of cyclical factors on repairable systems, we utilized the BEAST time series decomposition analysis. This allowed us to differentiate between abrupt changes, seasonal variations, and trends in the population of repairable systems. The effectiveness of this approach was confirmed through a real-life case study. The study offers a robust approach for analyzing functional degradation, thereby enhancing the understanding of degradation patterns.

 
Keywords Degradation pattern, variation, time series, functional clustering
   
    Article #:  RQD2024-229
 

Proceedings of 29th ISSAT International Conference on Reliability & Quality in Design
August 8-10, 2024