When nuclear power plant (NPPs) is in fault, it may release radioactivity into the environment. Therefore, extremely high safety standards specification are required during its working. So it is critically important for fault detection and diagnosis (FDD). NPPs are composed of large and complex systems, it is of great significance to obtain the up-to-date information of NPPs’ running state. So FDD is used to provide the state of system accurately and timely in NPPs. Signed directed graph (SDG) can show the complex relationship between parameters and has advantages of conveniently modeling, flexible inference and so on, so SDG is adopted for FDD. To achieve SDG inference better, fuzzy theory is utilized for signal processing in the paper. Firstly, SDG model is built according to the basic steps and principles of SDG modeling, and the parameters are divided into three states which is monitored by fuzzy theory. Secondly, according to the status of parameters, SDG is used for FDD and to reveal the fault propagation path, thus possibility of each fault occurred is achieved. Finally, to verify the validity of the method, the simulation experiments are done for NPPs and the simulation experiments show that SDG-fuzzy theory framework for FDD can get the fault possibility and deeply explain the reasons of fault.
A Fault Identification Method in Distillation Process Based on Dynamic Mechanism Analysis and Signed Directed Graph
