Research On Multi-spectral Radiation Thermometry Technology Based On Infrared Spectrometer

Temperature measurement plays an extremely important role in military defense,industrial production and epidemic prevention and control.With the development and progress of society,the difficulty and requirements of temperature measurement are getting higher and higher.Based on the actual needs,this paper develops a set of versatile portable multi-spectral radiation thermometers,researches on the identification method of multi-spectral radiation temperature measurement theoretical emissivity model,and achieves true temperature measurement of multiple targets.This article mainly starts from the design of the hardware part of the thermometer,the software,the instrument calibration and the target emissivity model identification method.The hardware part of the thermometer uses a customized infrared lens to collect the target radiant energy,and transmits it to the interior of the thermometer through the infrared hollow fiber.The spectroscopic system adopts a symmetrical Czerny-Turner optical path structure based on blazed grating spectrometry,uses a Pb Se focal plane detector array to receive spectral radiation,and is connected to the host computer through a USB interface.The thermometer software part runs on the PC upper computer,and communicates with the lower computer through the USB interface.The software design is developed based on the Lab VIEW graphical programming language,providing users with an intuitive and easy-to-use human-computer interaction interface.The software implements functions such as control,data acquisition,data processing,and data display of the lower computer.The overall size of the instrument is small,so it has good portability.The calibration of the instrument is divided into two parts: wavelength calibration and temperature calibration.The wavelength calibration uses the laser calibration method to determine the number of wavelengths corresponding to all channels;the temperature calibration uses the entire Baidu calibration method,and the least-squares method is used to complete the third-fit relationship parameter between the intensity signal of each channel and the bright temperature temperature value.Finally,the data modeling and classification methods of five typical emissivity models based on brightness and temperature spectrum are designed and demonstrated.The comprehensive recognition rate of theoretical simulation reaches 94.78%.After determining the emissivity model of the measured target,based on the channel data filtered by the threshold,a multi-spectral radiation temperature measurement model based on brightness temperature is established,and the final emissivity coefficient and real temperature are solved using a Simulated Annealing algorithm.