Process of threshold of fault diagnosis expert system in LAN based on BP neural network

Traditional control system usually uses a single expert system for fault-diagnosis, which is unitary and unable to deal with problems in the complicated and flexible network environment. In this paper, expert system and neural network system are combined for fault detection and diagnosis, which can greatly increase the efficiency and accuracy of network fault diagnosis. The networked control system discussed in the paper have several indices for fault diagnosis, which can be processed as input of BP neural network, and the training result proved to be good.