基于差异信息量的多源数据融合方法

D-S证据理论可应用于多源数据融合领域，但在处理高度冲突的证据时，可能会出现反直觉的结果.为解决这一问题，本文提出了差异信息量的概念及融合方法.首先，通过信息熵表明证据的相对重要性，采用散度获取证据可信度.然后利用证据可信度优化证据差异度以得到差异信息量，经过计算获取数据的最终权重，并将其作为D-S证据理论中的基本概率分配进行决策.在处理冲突证据、一致证据及不同数量证据等方面的数据融合问题时与其他方法对比，所提方法收敛更快，准确度更高.故障诊断的应用实例表明，所提方法的不确定性更小，优于现存的其他方法.

Multi-source Data Fusion Method Based on Difference Information

D-S evidence theory can be applied to the field of multi-source data fusion. However， counter-intuitive results may come out when handing highly conflicting evidences. In order to solve this problem， a modified fusion method with the concept of difference information(DI)was proposed. First， information entropy indicated the relative importance of evidence， and divergence was used to obtain the credibility of evidence. Then， the evidence difference was optimized by the credibility of the evidence to obtain difference information(DI). The final weight of the calculated data was used as the basic probability distribution in D-S evidence theory for decision-making. Compared with other methods in dealing with conflicting evidence， consistent evidence， and different amounts of evidence， the proposed method converges faster and has higher accuracy. The application examples of fault diagnosis show that the proposed method has less uncertainty and is better than other existing methods.