基于RS-FNN的核电厂设备智能故障诊断方法的研究

将粗糙集(RS)理论与模糊神经网络(FNN)相结合,能充分发挥各自的优点.本文利用RS方法对知识的约简技术,从大量的原始数据中提取精简的规则,基于这些规则建立的FNN网络具有更好的拓扑结构,学习速度大大提高、判断准确、容错能力强,具有更高的实用价值.为了验证该方法的有效性,以核电厂设备蒸汽发生器U形管破裂等故障为例,进行了仿真实验研究.诊断结果表明,将基于RS理论的FNN智能故障诊断方法引入核电厂设备故障诊断中是可行的,并且具有简单方便、计算量小、诊断结果可靠等特点.

Study on Intelligence Fault Diagnosis Method for Nuclear Power Plant Equipment Based on Rough Set and Fuzzy Neural Network

Rough set theory and fuzzy neural network are combined, to take full advantages of the two of them. Based on the reduction technology to knowledge of Rough set method, and by drawing the simple rule from a large number of initial data, the fuzzy neural network was set up, which was with better topological structure, improved study speed, accurate judgment, strong fault-tolerant ability, and more practical. In order to test the validity of the method, the inverted U-tubes break accident of Steam Generator and etc are used as examples, and many simulation experiments are performed. The test result shows that it is feasible to incorporate the fault intelligence diagnosis method based on rough set and fuzzy neural network in the nuclear power plant equipment, and the method is simple and convenience, with small calculation amount and reliable result.