Skip to content

Categories

On this page

2023-02-15T00:00:00.000Z

On this page

2023-02-15

Paper 01: Mechanical fault diagnosis and prediction in IoT based on multi-source sensing data fusion

huangMechanicalFaultDiagnosis2020 link DOI

Keywords: data-fusion, multi-sensor, fault-detection

Data Fusion Model

Data fusion model:

  • JDL model,
  • Boyd’s OODA (Observe-Orient-Decide-Act) Loop model,
  • Intelligence Cycle model,
  • the Omnibus model,
  • State Transition Data Fusion (STDF) model,
  • Hierarchical model
  • Dasarathy model,
  • Waterfall model,
  • Hybrid model.

Data Fusion Algorithm

  • simple processing algorithms
    • the weighted average method (weighting the average data acquired by the same type of sensors in continuous time series according to the chronological order, yielding the final fusion value).
  • probability-based algorithms,
    • Bayesian theory
      • Bayesian networks
      • naive Bayes
      • non-naive Bayes
      • Bayesian inference
    • D-S evidence theory.
      • preprocessing the original evidence
      • improvement of combination rules
      • combination of other algorithms
  • fuzzy logic-based algorithms,
    • fuzzy logic algorithm
      • fuzzy logic reasoning,
      • fuzzy clustering algorithm
      • intuitionistic fuzzy set
      • a combination of other algorithms
    • fuzzy set
    • rough set: attribute reduction and noise removal
  • artificial intelligence algorithms
    • neural networks
      • Back-Propagation Neural Network (BPNN)
      • Radial Basis Function Neural Network (RBFNN)
      • Elman Neural Network (ENN)
    • Support Vector Machine (SVM) for small number of samples
    • deep learning.
      • Convolutional Neural Network (CNN)
      • Deep Belief Network (DBN)
      • Deep Neural Network (DNN)
  • other algorithms.
    • ant colony algorithm
    • genetic algorithm
    • Kalman filtering

Fusion strategies

  • select a reasonable data preprocessing method
  • select a reasonable neural network model, and use a heuristic algorithm to optimize the network model.

Paper 02: Hierarchical attention graph convolutional network to fuse multi-sensor signals for remaining useful life prediction

liHierarchicalAttentionGraph2021 link DOI
Keywords: multi-sensor, RUL, graph

Tags

Edit this page
Last updated on 3/7/2023