On the Dempster-Shafer framework and new combination rules
Information Sciences: an International Journal
Default reasoning and possibility theory
Artificial Intelligence
The Combination of Evidence in the Transferable Belief Model
IEEE Transactions on Pattern Analysis and Machine Intelligence
Combining belief functions when evidence conflicts
Decision Support Systems
Combining belief functions based on distance of evidence
Decision Support Systems
Expert Systems with Applications: An International Journal
The Framework for Multi Agent Concurrent Negotiation
WMWA '09 Proceedings of the 2009 Second Pacific-Asia Conference on Web Mining and Web-based Application
Fault diagnosis in railway track circuits using Dempster-Shafer classifier fusion
Engineering Applications of Artificial Intelligence
Car assembly line fault diagnosis based on modified support vector classifier machine
Expert Systems with Applications: An International Journal
A Theory of Evidence-based method for assessing frequent patterns
Expert Systems with Applications: An International Journal
Journal of Network and Computer Applications
Hi-index | 12.05 |
Multi sensors fusion is a very important process for fault diagnosis system. Information obtained from multi sensors need to be fused because no single sensor can get all the information for fault diagnosis. Moreover, information from different sensors may be uncertainty, inaccuracy, or even conflicting. Evidence theory can be used for information fusion, which is regarded as an extension form of Bayesian reasoning, but it has a better fusion result by simple reasoning process using belief function without knowing the prior probability. All the information collected from multi sensors in the system can be described as the evidence for diagnosis so that the fault diagnosis problem can then be modeled as a problem of evidence fusion and decision. In this paper, the classical Dempster-Shafer evidence theory is discussed, and the disadvantages of the combination rule are also analyzed. The notion of support degree of focal element is suggested in order to evaluate the conflicts between multi sensors. The new combination rule is then built to allocate the conflicted information from multi sensors based on the support degree of focal element. Furthermore, the decision rules for fault diagnosis are also proposed, as well as the architecture of the agent oriented intelligent fault diagnosis system. Finally, a case study is given to illustrate the performance of the proposed model.