Sign in Subscribe
Northeastern University

Distributed Fault Detection for Network Systems: A Novel Approach

Strategic Insights

— 1 min read

A team of researchers from Northeastern University has made a breakthrough in developing a novel approach to fault detection for network systems composed of multiple clusters with unknown system matrices. This approach, which utilizes the subspace instrumental variable method and unknown input decomposition, has been shown to be effective in detecting faults in local cluster systems.

Key Takeaways:

  • The proposed method, called distributed fault detection, utilizes the intersection of subspaces on local observations as the states of connections to identify local cluster subsystem matrices.
  • The subspace instrumental variable method is used to design an unknown input observer (UIO) to detect faults in local cluster systems.
  • The unknown input decomposition approach is presented to address the rank conditions for designing UIO, eliminating the decouplable part and attenuating the impact of the undecouplable part using a robust performance index.
  • The effectiveness of the proposed fault detection scheme is verified via numerical simulation and comparative analysis.
  • The research has been funded by the National Natural Science Foundation of China (NSFC), Aeronautical Science Foundation of China, and Research Fund of State Key Laboratory of Synthetical Automation for Process Industries.
  • The method is suitable for network systems with unmeasurable connections between clusters.

Xiao-Jian Li, the lead researcher from Northeastern University, has developed this novel approach, which has the potential to improve the reliability and performance of network systems.

Statistics:

  • 150 (current issue of Communications In Nonlinear Science and Numerical Simulation)
  • 2025 (year of publication of the research)
  • 108 (page number of the research article in Mathematics Week)
  • 1 research article (Distributed Fault Detection for a Class of Network Systems: Optimal Unknown Input Observer Design) published in Communications In Nonlinear Science and Numerical Simulation
  • 4 funders (National Natural Science Foundation of China, Aeronautical Science Foundation of China, Research Fund of State Key Laboratory of Synthetical Automation for Process Industries)

Sources:

  • Distributed Fault Detection for a Class of Network Systems: Optimal Unknown Input Observer Design, Communications In Nonlinear Science and Numerical Simulation, 2025; 150
  • Xiao-Jian Li, Northeastern University, College of Information Science and Engineering, Shenyang 110819, People's Republic of China
  • National Natural Science Foundation of China (NSFC)
  • Aeronautical Science Foundation of China
  • Research Fund of State Key Laboratory of Synthetical Automation for Process Industries

Read more