Title :
An improved combination rule in fault diagnosis based on dempster shafer theory
Author :
Deng, Yong ; Wang, Dong ; Li, Qi
Author_Institution :
Sch. of Electron. & Inf. Technol., Shanghai Jiao Tong Univ., Shanghai
Abstract :
In this paper, a modified method of conventional DS evidence theory is proposed. The consensus index of evidence based on distance measures of evidence is introduced into modifying the BPAs of hypotheses with the evidence sufficiency and evidence importance. By using the Dempsterpsilas combination rule and new decision rules, the different sets of BPAs for each hypothesis are calculated, respectively, and the result can be easily obtained. A numerical example of fault diagnosis is used to illustrate the proposed method and compare with other methods.
Keywords :
fault diagnosis; inference mechanisms; Dempster Shafer theory; combination rule; decision rules; evidence theory; fault diagnosis; Bayesian methods; Decision making; Fault diagnosis; Hydrogen; Information technology; Machine learning; Robustness; Uncertainty; Conflict; DS evidence theory; Evidence consensus;
Conference_Titel :
Machine Learning and Cybernetics, 2008 International Conference on
Conference_Location :
Kunming
Print_ISBN :
978-1-4244-2095-7
Electronic_ISBN :
978-1-4244-2096-4
DOI :
10.1109/ICMLC.2008.4620406