Research On Real-Time Detection System And The Methods Of Fault Diagnosis For The Electric Equipments

In the past few years, with the rapid development and application of computer, integrated circuit, signal process, communication and optical fiber technology and the gradual improvement of sensing probe and other electrical power unit, the development and application of Real-Time Detection and Control System has come true and been improved gradually. This thesis will do some research on the methods of fault diagnosis based on the modeling and fuzzy neural network. The electric power system is divided into several levels. The fault diagnosis on the daughter systems will make full use of the advantage of fault diagnosis, as the fault concept is easier to command compared with the whole system and the mathematical models are easier to build, especially that the modeling problems of every type of technology process have been worked out in the past few years. The results in this field have been used completely in fault diagnosis. Upon the research on the whole electric power system, the robustness of fault diagnosis algorism is bad and it is hard to build practical math models for some objects. For these diagnosis problems, this thesis will research the expert system based on the combination of neural network and fuzzy logic system in the Real-Time System, in which fuzzy variables are used. Using this method will make full use of experts'experience and the system will have fault-tolerant ability.