A New Evidential Reasoning-Based Method for Online Safety Assessment of Complex Systems

Fu Jun Zhao, Zhi Jie Zhou, Chang Hua Hu, Lei Lei Chang, Zhi Guo Zhou, Gai Ling Li

Research output: Contribution to journalArticle

27 Scopus citations

Abstract

It is vital to online assess the safety of a complex dynamic system by taking into account the current state, degradation trend, and historical records together. This paper proposes a new safety assessment model with an online algorithm based on the evidential reasoning (ER) approach. It does not only take into account the relative importance of each safety indicator, but also consider the reliability of each indicator. To obtain the integrated safety level, multiple safety indicators are fused at first and the 'history,' 'current,' and 'future' safety states are then integrated. First, a forecasting model based on the third-order Volterra filter is proposed to predict the safety indicators' information online. Second, an adaptive weighting model is developed to automatically adjust to various conditions and track the characteristics of the dynamic system, and the reliability of each indicator is considered to diminish the influence of inherent disturbance and/or noise. Finally, a safety assessment aggregation scheme based on the ER approach is presented to fuse the history, current, and future safety indicators to obtain the corresponding safety state, and the safety states are then fused synthetically to obtain a comprehensive safety assessment result of the complex dynamic system. A numerical study is examined to demonstrate the implementation and effectiveness. Moreover, a practical example of the inertial navigation system is studied to show the potential applications of the proposed ER-based safety assessment method.

Original languageEnglish (US)
Pages (from-to)954-966
Number of pages13
JournalIEEE Transactions on Systems, Man, and Cybernetics: Systems
Volume48
Issue number6
DOIs
Publication statusPublished - Jun 1 2018

    Fingerprint

Keywords

  • Evidential reasoning (ER)
  • information fusion
  • online safety assessment
  • reliability
  • volterra filter
  • weight

ASJC Scopus subject areas

  • Software
  • Control and Systems Engineering
  • Human-Computer Interaction
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this