基于模糊神经网络的复杂网络故障诊断设计

针对雷达等复杂大型电子装备网络系统的故障定位难、影响关系不清晰的问题，采用了基于模糊神经网络的故障定位方法，提高了网络故障定位的快速性与准确性：首先介绍了模糊隶属度及模糊神经元等理论，接着基于模糊理论将网络监测信息进行模糊化处理，并利用神经网络模型对模糊后的信息进行训练与学习，参数训练达到设置的期望误差0.01；最后利用训练好的模型对随机抽取的2组网络故障实例进行了验证，软件执行单次诊断耗时3.5s；结果表明采用基于模糊神经网络的诊断方法，能够较好解决网络故障耦合复杂、故障现象与故障原因关系不清晰等难题，对网络故障快速排除与恢复具有重要意义。

Design of Fault Diagnosis for Complex Network Based on Fuzzy Neural Network

Aiming at the problems of difficult fault location and unclear influence relationship in complex large-scale electronic equipment network systems such as radar, a fault location method based on fuzzy neural network is adopted, which improves the speed and accuracy of network fault location. Firstly, the theory of fuzzy membership degree and fuzzy neuron is introduced, then the network monitoring information is fuzzified based on fuzzy theory, and the fuzzy information is trained and learned by using the neural network model. And the parameter training reach the set expected error of 0.01.Finally, the trained model is used to verify two groups of randomly selected network failure instances. It takes 3.5s for the software to perform a single diagnosis. The results show that the diagnosis method based on fuzzy neural network can better solve the problems of complex network fault coupling and unclear relationship between fault phenomenon and fault cause, which is of great significance to the rapid elimination and recovery of network faults.