Diagnosis and Analysis of Uncertain Information based on Probability-box
Yan Liu, Qing Xin Wang, Jia Man Ding · 2015
The fault information always contains many uncertain, such as the missing information, malfunction and failure to operation.In this paper, we use a new method based on probability-box theory and support vector machine to solve the uncertain problems and improve the diagnosis ability.The first, structural the p-boxes based on the data of fault record.The second, using the fusion rules to get the different probability boxes fusion, then extract the feature from the p-boxes.The final, get the diagnosis result based on the support vector machine.This paper shows that the probability box has high diagnostic rate by comparing with the traditional method.