• DocumentCode
    3573947
  • Title

    Multi-interpretations Valued Tolerance Relation model of incomplete information in fault diagnosis system

  • Author

    Xiaojun Lu ; Wentao Huang ; Weijie Wang ; Yu Jun

  • Author_Institution
    Dept. of Mechatron. Eng., Harbin Inst. of Technol., Harbin, China
  • fYear
    2014
  • Firstpage
    6035
  • Lastpage
    6040
  • Abstract
    Because of the complex structure and terrible running environment of mechanical equipment, in some extent, the fault diagnosis information obtained from the mechanical fault diagnosis is incomplete. At present, the processing of the incomplete fault diagnosis information is carried out by replacing the unknown attribute values with the known attribute values using the Normal Valued Tolerance Relation. Although this kind of method can obtain diagnostic information, it cannot match the multiple reasons that cause the incompletion of fault diagnosis information. In this paper, the reasons that cause the incompletion of fault diagnosis information is studied and the incomplete information in multiple semantics is explained, and the Multi-interpretations Valued Tolerance Relation is proposed according to the features of mechanical fault diagnosis. The relation is verified to be correct by calculation and analysis of several examples. The relation proposed in this paper can provide theoretical basis for the knowledge acquisition in fault diagnosis.
  • Keywords
    fault diagnosis; maintenance engineering; complex structure; fault diagnosis information; fault diagnosis system; knowledge acquisition; mechanical equipment; mechanical fault diagnosis; multiinterpretations valued tolerance relation model; normal valued tolerance relation; Fault diagnosis; Information systems; Lead; Media; Probability distribution; Semantics; Set theory; fault diagnosis; incomplete information; multi-interpretations; relation model;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Intelligent Control and Automation (WCICA), 2014 11th World Congress on
  • Type

    conf

  • DOI
    10.1109/WCICA.2014.7053754
  • Filename
    7053754