基于粗糙集知识获取的转子轴承系统故障诊断知识库构造研究

利用粗糙集理论从信息系统中知识获取与知识发现的功能，对转子轴承系统状态监测与故障诊断系统中知识库构造进行了研究。通过对粗糙集理论用于智能决策过程和诊断监测数据特性的分析，获得了数据表中连续属性值离散化映射、条件属性集与决策属性集记录的完备性是诊断知识获取与知识发现前提条件的结论。建立了基于粗糙集方法从数据库中知识获取和知识发现的诊断知识库模型，设置的连续属性值离散化步骤具有实现知识库决簏规则扩展的特点。

Study on Building a Fault Diagnosis Knowledge Base of Rotor-bearings System Based on Knowledge Acquisition of Rough Set

By analyzing the intelligent decision procedure based on Rough Set and comparing with the parameters obtained by condition monitoring of rotor bearings system,the special characteristics of a model of fault diagnosis knowledge base were extracted.One characteristic is that the preconditions of knowledge acquisition and knowledge discovery from a database is to discrete the continuous attributes values into special integer numbers or enumerate variable.Another is that a record about the condition attributes set and decision attribute set must be completive.So,the knowledge base model of knowledge acquisition and knowledge discovery based on rough sets were built in condition monitoring and fault diagnosis device of rotor bearings system.It emphasizes the process of discretization and also sets the applicable procedure.The model has the capability to extend the decision making rules of the knowledge base.