中文摘要：

摩擦学状态辨识实质上是分类问题，针对以往机器摩擦学状态判别主要依靠人工经验来完成所存在的缺陷，用知识发现的思想来解决摩擦学状态辨识的知识获取问题。采用加权ID3（iterative dichotomizer 3）算法来度量摩擦学状态监测实例表中各条件属性对状态辨识的重要性，建立了基于摩擦学状态监测实例库和决策树的知识获取方法模型。将模型应用于磨损试验的摩擦学状态辨识的知识挖掘分析，利用获得的知识对测试集进行状态识别，取得了良好的摩擦学状态辨识结果，从而为从监测实例中挖掘摩擦学状态辨识知识提供了方法与手段支持。

英文摘要：

Tribological systemic condition identification is virtually a problem of classification. In order to solve the problem of knowledge acquisition in tribological systemic condition identification which mainly depends on human＇s experiencenowadays, the idea of knowledge discovery was adopted herein. ID3 was used to measure the priority of condition attribute to decision trees in tribological systemic condition monitoring. Then, the knowledge acquisition model was established based on the example data base of tribological systemic condition monitoring and decision trees. Finally, the developed knowledge acquisition model was validated in knowledge mining analysis of tribological systemic condition identification based on wear tests. The example application demonstrates that the developed model provides the good solution in the knowledge acquisition of tribological systemic condition identification.