The Research Of Diagnosis Instrument For Railway Based On Frequency Spectral Analysis And Fuzzy Neural Network

Since 201L century, the society of human beings has stepped into the era of Industrial Revolution . Now the production is more and more concentrated, continual and precise. So there rise some questions: how to make the equipment work properly; If there axe faults on the equipment, how could we locate and debug them? All the questions above are the practical and industrial background for the theory of fault diagnosis. The science develops at a very high speed. At the same time the manufacture has been very precise. Both stimulate and spur the development of the theory. This is the theoretical and technical background for the theory of fault diagnosis. Now the theory of fault diagnosis has been widely used in many fields. It has brought about much economic effect which is the reason that the industrial divisions of all countries attach importance to it. In modem society railroad has play an unique and important role. The subway of the metropolis and the transportation of the goods depend on the railroad. If there are some faults on the railroad ,it may effect the run of the subway and the transportation of the goods. So the status of the railroad must be pparent?to us. If the railroad does not run properly ,how could we iagnose?these faults ? Via the equipment that we has invented, it is very easy for us to know the status of the railroad. The research on the fault diagnosis for the railway is based on the spectral analysis and fi.izzy neural network. Chapter one involves the general knowledge and background of fault diagnosis. Chapter two involves the faults sources . Chapter three involves the development of the diagnosis instruments. Chapter four and five which are the core of the paper involve the knowledge of spectral analysis ~. fuzzy theory~ neural network and their applications. Chapter six involves the realization of the instrument.