蒸汽发生器U形管破裂事故的识别与诊断系统

Recognition and fault diagnosis system of U shape pipe breaking accident of steam generator

论述了船用核动力装置中蒸汽发生器U形管发生破裂时的故障征兆及对U形管进行故障诊断的必要性,阐述了智能控制领域中的两个方面即模糊逻辑和神经网络,并探讨了它们之间的结合—模糊神经网络结构及其实现算法,利用模糊神经网络对蒸汽发生器U形管破裂事故进行了诊断,诊断结果表明该理论方法对此事故完全可以正确识别,进而证明该理论方法可以应用到船用核动力装置其他故障的诊断,能够满足船用核动力装置的诊断要求.

The fault symptoms of the U shape pipe breaking accident of steam generator (USPBASG) and the necessary for diagnosing the USPBASG are discussed. Meanwhile, the differences and connections between the fuzzy logic and neural network are compared. Furthermore, the algorithm and structure of the fuzzy neural network are introduced and applied to the Fault Diagnosis for the USPBASG. The test result shows that the fault can be identified by the fuzzy neural network. Moreover, the fuzzy neural network can diagnose other faults of the nuclear power plant, and it can satisfy the demand for fault diagnosis.