A new correlation coefficient of mass function in evidence theory and its application in fault diagnosis

摘要

Evidence theory as a generalization of probability theory has an advantage of handling uncertainty. The mass function, also named as basic probability assignment (BPA), plays an important role in evidence theory. Many kinds of measures were proposed based on mass functions like entropy, distance, negation. However, how to determine the correlation coefficient between two mass functions is still an open issue. In this paper, a new correlation coefficient is proposed, which can better show the relationship between mass functions. In numerical examples, the effectiveness of proposed coefficient is illustrated by comparing with existing methods. Besides, an application of fault diagnosis is given.