| The multispectral temperature measurement method has its uniqueness in the field of true temperature measurement of the high temperature.This method constructs the true temperature and emissivity of the target by measuring the radiance information in multiple spectrums of the target at the same time.It is a true temperature measurement method for high-temperature targets with high measurement accuracy,wide temperature measurement range,and wide scope of application.It has been successfully applied to the solid rocket motor plume true temperature measurement,explosion flame true temperature measurement and many other occasions.With the increasing demand for temperature measurement,the target of the high temperature measurement has gradually changed from the single target point and multiple target points to the two-dimensional true temperature field.However,there are currently few multispectral pyrometers that can measure the true temperature field of the two-dimensional target.The difficulties of the two-dimensional true temperature field measurement technology mainly include the following points:proposing a fast multispectral true temperature inversion method,developing a two-dimensional true temperature field measurement system,proposing an intelligent judgment method for optical filter array assembly of the two-dimensional true temperature field measurement system.Aiming at the above difficult problems,this paper carries out the relevant research work.The main research contents of this paper are as follows:(1)A fast multispectral true temperature inversion method based on emissivity constraints is proposed.The traditional multispectral temperature measurement method is proposed for the true temperature measurement of a single target point or multiple target points.It requires complex operations such as iteration to obtain the true temperature and emissivity of the target.When measuring the true temperature field of a two-dimensional target,the number of measurement targets will grow exponentially.If the traditional multispectral temperature measurement method is used again,the problem such as too long time for the true temperature field inversion will arise.This issue will seriously affect the practicality of the two-dimensional true temperature field measurement system.This paper reveals the internal relationship between radiant energy equivalent and emissivity on the theoretical level,and find the conditions that constrain target emissivity based on target radiation data,which can more fully use target radiation information to limit the construction of emissivity model.This work eliminates a large number of redundant samples and provides the theoretical basis for the selection and reduction of emissivity samples,which can improve the speed without reducing the accuracy of multispectral true temperature inversion.This paper verifies the effectiveness of this method through simulation experiments,and the results show that the true temperature calculation time is saved by 12% to 38%.(2)A two-dimensional true temperature field measurement system based on the combination of multipoint multispectral lateral detection and rotating mirror longitudinal scanning is developed.The radiation temperature field of the measured target can be measured by using thermal imager,CCD(Charge-Coupled Device)and other radiation thermometers.But it is difficult to measure the true temperature field because of the unknown emissivity.At present,multispectral thermometers that can measure true temperature are mainly based on single or multiple target points.There have been few reports on instruments for measuring the true temperature of the twodimensional field.This paper focuses on the development of the two-dimensional true temperature field measurement system from three aspects: optical path,circuit,and software.In the optical part,an optical probe with focusing function is used,and the radiation information of the temperature field is obtained by combining multipoint multispectral lateral detection with longitudinal scanning of a rotating mirror to balance hardware resources and measurement speed.The circuit part adopts the design scheme of the multi-board cascade to achieve a large amount of data at the same time.The software part adopts a multi-threaded parallel processing method to improve computational efficiency,which reduces the time by 77% compared with single threaded processing.(3)An intelligent judgment method for optical filter array assembly of the twodimensional true temperature field measurement system based on data sequence recognition is proposed.The traditional PWF(Pyrometer Wavelength Function)method for wavelength calibration requires specialized calibration devices such as the monochromator.This method has multiple experimental steps,long operation time,and requires manual calibration.It is not suitable for measuring the instruments equipped with a large number of filters,such as the two-dimensional true temperature field measurement system.This paper finds the relationship between the change of blackbody radiation energy equivalent and the change trend of wavelength from the data sequence by revealing the internal relationship theoretically,proposing a new method for intelligent judgment of all filter assemblies by using only the instrument temperature calibration data.This paper verifies the effectiveness of this method through simulation experiments.(4)Three actual experimental experiments are conducted,including the validation experiment of the fast multispectral true temperature inversion method,the temperature measurement experiment of the two-dimensional true temperature field measurement system,and the validation experiment of the intelligent judgment method for filter array assembly.The uncertainty analysis of true temperature measurement is also conducted.The experimental results show that the fast multispectral true temperature inversion method saves 19% to 27% of calculation time without reducing accuracy.The measurement system can achieve the true temperature measurement data of the two-dimensional field.The effectiveness of the intelligent judgment method for filter array assembly is verified.The expanded uncertainty of true temperature measurement is 2.46%. |
