为有效而充分地利用故障信息以提高诊断精度,本文受生物感知系统“先分后合”信息处理方式的启发,提出了故障多征兆域集成诊断的诊断思想,并对该思想的实现途径进行了研究,提出了基于模糊积分理论实现集成处理的可行策略,最后就所提出的方法和策略进行了示例研究。%To efficiently make use of more fault information to improve diagnosis precision, an idea of fault multi-symptom-domain integrated diagnosis is developed, which achieves fault diagnosing in various symptom domains respectively, then makes consensus diagnosis decision among all symptom domains. From the point of view of group decision, the methods based on neural networks to realize this diagnosis idea are studied, and a particular multi-symptom-domain integrated diagnosis strategy is proposed, which uses fuzzy integrals theory to integrate the diagnosis results from the neural networks in various symptom domains. Finally, a case study about rotor system fault diagnosis is given. The research results show that the proposed diagnosis strategy is feasible and more efficient than conventional methods.
展开▼