RUL prediction for multi-component hierarchical systems with external environmental effects and stochastic dependence of multiple indicators

Authors:
Publisher website
Google Scholar
Abstract
The advances in technology in the manufacturing industry has enabled the construction of intricate systems comprising multiple key components. The degradation of these components can usually be characterized based on multiple indicators. The correlation between multiple indicators of any component and the interdependence between components cannot be ignored. Moreover, time-varying loads and dynamic conditions exert additional random effects on the system degradation process, which affects the remaining useful life (RUL) prediction. Thus, in an attempt to overcome these issues, this study proposes a hierarchical system RUL prediction method under the influence of the external environment that considers two different kinds of the random correlations among multiple indicators of any component and among multiple components. First, multiple indicators representing the degradation of any component are constructed using multiple features. A distributed fusion degradation modeling is performed by considering the stochastic correlation among the indicators of a component. Second, considering the stochastic relevance among multiple components and the influence of measurable and unmeasurable external environments, the degradation model of the multi-component system is constructed. Meanwhile, the parameters in the model are estimated and updated step-by-step. Furthermore, the RUL of a multi-component system with different structural types is predicted. Finally, the effectiveness of the proposed method is verified based on numerical examples and practical cases. Compared with the existing methods, the root mean square error of this method is reduced by 17%, and the SF is increased by 19%. This model can more intuitively reflect the hierarchical random correlation between complex multi-component system degradations.
Funding Information
  • Scientific and Technological Achievements Transformation and Guidance Project of Shanxi Province (202404021301040)
  • Scientific and Technological Innovation Programs of Higher Education Institutions in Shanxi (2025L122)
  • Key Research and Development Projects in Shanxi Province (202202090301011)
  • Taiyuan Institute of Technology Scientific Research Initial Funding (2024KJ040)
  • Excellent Doctor Award Fund for Working in Shanxi Province (2025LJ023)
  • National Natural Science Foundation of China (61703297)
  • Major Science and Technology Project of Shanxi Province (202201090301013)

This publication has 38 references indexed in Scilit: