集成旋转机械故障诊断的知识表示与推理

An Integrated Knowledge Representation and Inference for Rotating Machinery Fault Diagnosis

本文提出了一种新的基于规则和神经网络集成的智能旋转机械故障诊断方法，该法把专家经验和故障样本以统一的分布表示形式组织到知识库中，并在此基础上提出了推理算法的自学习选取。整个系统充分发挥了规则系统和神经网络的优点，具有知识表示明确、并行推理、联想和自学习等优点，最后结合实例进行了分析。

A new mintelligent rotating machinery fault diagnosis method which integrates rule-basedsystem with neural network is presented.Expert experience and samples are included into knowledgebase with uniformly distributed representation form. Adaption of inference method is aiso presented. The system has advantages of both rule-based system and neural network, such as:distinet knowledgerepresentation, paralled uiference, generalization and self-learning.As example is presented in theend.