Open AccessInformation Fusion of Discrete Variables and Continuous Variables Based on D-S Evidential Theory
Author(s) -
Huicheng Yang,
Hong-Mei Sun,
H. X. Chen,
GuangJun Jiang,
C. W. Li
Publication year - 2021
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/1043/2/022028
Subject(s) - entropy (arrow of time) , dempster–shafer theory , discretization , computer science , continuous variable , data mining , reliability (semiconductor) , probability distribution , variable (mathematics) , mathematics , mathematical optimization , statistics , mathematical analysis , power (physics) , physics , quantum mechanics
In reliability analysis of a target, information from various aspects should be analyzed comprehensively. In this process, information fusion plays a key role in integrating and analyzing information from different channels. Aiming at the reliability analysis of discrete variables and continuous variables existing simultaneously, an information fusion method based on Dempster’s rule of combination of D-S evidence theory was proposed to facilitate the reliability analysis by integrating information. Firstly, the continuous variables are discrete by using the continuous variables discretization algorithm based on the probability distribution of the job state with equal expectation scale, and the basic probability distribution of each discrete variable is carried out. Then Dempster’s synthesis rule was used to synthesize the discrete variables. Deng entropy was used as the judgment standard in the synthesis process, so that the entropy of each step of fusion was reduced to ensure the credibility of fusion was increased. Finally, the feasibility and effectiveness of the method are verified by a case.