Research Of Datafor Train Hotbox Detection System Based Ona Multi-point Infrared Detector

Railway infrared axle temperature detection system is a real-time system tomonitor the axis temperature of trains and plays an important role in avoiding train’saccidents. In recent years, with the progress of technology and train’s speed up, thesystem does not meet higher and more strict requirements for train’s running safety.But due to various reasons, monitoring system is still not ideal in detecting hot axles.Because it is not only related to the hardware, but also related to computer processingtechnology, which is also the means to improve efficiency. So, for the new axlestemperature detection system, how to use computer points to deal with a lot of hot boxtemperature data is effectively improve somewhat better than the key technicalproblem.According to the present stage has completed the line formation infrared axletemperature detection system research and the current research requirements, thispaper mainly has been done three aspects of the work. First is the computerprocessing of data fusion technology optimization research, Second is the computerprocessing technology of artificial intelligence optimization research, the last is thedata mining technology in the field of temperature data mining.Specific work is asfollows:The first, it illustrates the significance of the research and the background, andthe present stage has been completed for line formation infrared axle temperaturedetecting system.The second, research and analysis of temperature data. waveform of axletemperature, the abnormal wave, the influence of the speed of axial temperature dataare analyzed, the system adopts the waveform and temperature data form wasanalyzed, and finally determine the data processing flow of the system.The third, temperature optimization of data fusion research. The computerprocessing of data fusion technology is optimized, the purpose is to improve theaccuracy and efficiency of data fusion.The fourth, the research of discriminant axle status. The computer processing ofartificial intelligence technology to optimize, purpose is to improve the efficiency ofthermal one axis and discriminant.The fifth, the preliminary research on the temperature data mining. It uses thedata mining technology in the axle temperature database, analyzing historicaltemperature data mining, and to predict hot axles.