基于定性模型和定量知识集成的智能故障诊断方法研究

Research of Intelligent Fault Diagnosis Method Based on Integration of Qualitative Model and Quantitative Knowledge

首先从诊断对象的知识表示和处理诊断知识的能力两个方面对基于定性模型的故障诊断方法进行了分析,然后在此基础上提出了基于深浅知识集成的分层诊断策略,并基于符号有向图和模糊理论提出了一种定性模型和定量知识集成的智能故障诊断方法。仿真故障数据验证结果表明,该方法不仅具有较快的诊断速度,而且能有效地提高诊断的准确度。

An analysis of qualitative model-based diagnosis methods is first proposed in the view of knowledge expression and management ability, then a hierarchy strategy of fault diagnosis is developed based on the integration of shallow and deep knowledge. An intelligent fault diagnosis method is also developed based on the integration of qualitative model and quantitative knowledge in the way of signed directed graph and fuzzy theory. Results tested with simulated fault data show that the method provided can diagnose faults quickly and improve the veracity of diagnosis effectively.