机器磨损状态监测中的证据理论方法

Evidence Theory Method in Wear Condition Monitoring

阐述了证据理论适应于决策融合的特点，并以磨损状态监测数据为基础，探讨了识别框架的形成、基本概率指派的构造以及相应证据的组合与推理思路，说明寻求简便有效的证据融合算法、研究与之相适应的数据库结构及信息融合过程的计算机实现是关键步骤之一。

The characteristic of evidence theory using for decision fusion is set forth. On the basis of data in wear condition monitoring, forming identification flame, constructing basic probability assignment and synthesizing and reasoning corresponding evidences are discussed, it is showed that seeking for a single and effective algorithm of evidence fusion, studying database construction and information fusion implementing in computer is one of key steps.