Dissertation > Excellent graduate degree dissertation topics show

Information Fusion Method Based on Desert-Smarandache Theory for Fault Diagnosis

Author: JiangHaiNa
Tutor: WenChengLin
School: Hangzhou University of Electronic Science and Technology
Course: Control Theory and Control Engineering
Keywords: Evidence Theory DSmT Frame of discernment Basic probability assignment Concurrent fault diagnosis
CLC: TH165.3
Type: Master's thesis
Year: 2009
Downloads: 95
Quote: 1
Read: Download Dissertation

Abstract


Dempster-Shafer theory (DST) is an information fusion method. Due to the representation of the uncertainty, the the metric combination with the advantage, DST has been in the field of fault diagnosis with a wide range of applications. However, due to the imperfection of the theoretical framework (such as strict requirements identification framework elements are mutually exclusive), often appeared to be inadequate in dealing with the concurrent fault diagnosis. Dezert-Smarandache theory (DSmT) proposed in recent years to break through the limit the recognition framework required elements mutually exclusive, and so be able to deal effectively with single and concurrent fault representation, uncertain information, metrics and fusion establish a unified framework. DSmT diagnose problems with many advantages, but it still can not get rid of the DST fault diagnosis related evidence handling, identification framework is not complete and dynamic change problem. To this end, this paper carried out the following research: (1) analysis of the DST in solving the evidence of correlation, identification framework is not complete and dynamic changes, the current domestic and international evidence theory and its improvement The methods were reviewed. (2) The proposed the concurrent information fusion method of fault diagnosis evidence. Characteristics of the concurrent fault, given that covers the identification of single and concurrent fault framework, as well as the DSmT fusion rule; different sensor failure characteristics and basic probability assignment method given new decorrelating method; Finally The fusion hybrid DSmT combination rule decorrelation evidence diagnostic results. (3) the establishment of to open recognition under the framework of concurrent information fusion method of fault diagnosis. The method can handle identification framework is not complete due to the conflict of evidence, extended DST in the two-dimensional conflict measure factor DSmT framework, with its judgment the level of conflict between the evidence and decided to adopt a closed or open to recognition under the framework of evidence combination rule. (4) to carry out the conditions of the the DSmT evidence combination method and its application in fault diagnosis. DSmT combination method handling limitations dynamically changing the frame of discernment; then given the DST framework under Dempster conditions rules and classes Dempster condition update rules of evidence; and DST under conditions rules promotion to the DSmT treated fault diagnosis . Also for (2) 4 to solve the problem of the experiment, respectively, in the multi-purpose motor flexible rotor experimental stage set, and verify the feasibility and effectiveness of the proposed method.

Related Dissertations

  1. Research on Joint Target Detection for Dual-Sensor Image and System Implementation,TP391.41
  2. Evidence-based cell model theory and semantic multi-tag music emotion recognition,TP391.41
  3. Water quality time series data processing and Early Warning System Construction Research Database,TP274
  4. Dynamic scheduling for steelmaking disturbance recognition and classification method,F426.31
  5. MIDI-based instrument control systems and automatic identification method notes,TN912.34
  6. Research on Early-warning and Forecasting System Algorithm for Distributor Fault Arc Based on Information Fusion,TM592
  7. The Rough Set Method Based on Evidence Theory,TP18
  8. Research on Route Planning Algorithm in Vehicle Navigation System,U495
  9. Design of Transient Signal Wavelet Analysis System and Study on Fault Line Selection of Resonant Grounded System,TM862
  10. Research and Implementation of Context Information Query, Subscription and Uncertainty Reasoning,TP391.1
  11. Researches of Spectrum Sensing Technology in Cognitive Radio,TN925
  12. Study on Transient Component Selective Protection Distribution Network Based on Wavelet Entropy and Evidence Theory,TM862
  13. Rolling Bearing Fault Diagnosis of Go Line Department in Railway Train,U270.7
  14. Data mining and information fusion in Coal Mine Safety Monitoring System Research,TP274
  15. Researches on Fusion Method Based on Multi-Channel Information of Sign Language,TP391.41
  16. The Research of Video Communication Protocol Fusion Based on D-S Evidence Theory,TN919.8
  17. Attribute Reduction Based on Extensions of Concepts in Formal Contexts,TP18
  18. The Similarity Measure Method of Conflicting Evidence and Its Application in Information Fusion Fault Diagnosis,TP202
  19. The Research on Decision Support System Based on Evidence Theory,TP182
  20. Research for Fault Diagnosis of Analog Circuits Based on Intelligent Information Fusion,TN710

CLC: > Industrial Technology > Machinery and Instrument Industry > Machinery Manufacturing Technology > Flexible manufacturing systems and flexible manufacturing cell > Fault diagnosis and maintenance
© 2012 www.DissertationTopic.Net  Mobile