Uncertainty Evaluation And Reliability Analysis Of High Precision Platinum Resistance Thermometers

Infrared remote sensing technology is commonly used in the fields of climate change,weather forecast,environmental monitoring and so on due to its advantages of high timeliness and high-precision in obtaining surface temperature.However,with the improvement of numerical weather prediction model,the field of temperature monitoring has put forward higher requirements for the measurement accuracy of infrared remote sensing load.The temperature calibration level will reach 0.1 ? in the next five to twenty years,while the infrared detector temperature calibration level of Fengyun satellites in China is currently 0.5 ? ? 1 ?.The shortage of the quantification level of infrared remote sensing load is the main reason for the bottleneck of high-precision measurement.In order to achieve the goal of high temperature and high water level calibration,a method of on-orbit calibration using calibration satellites is proposed.A high-precision space reference blackbody radiation source is installed on the satellite,and a specially designed high-precision platinum resistance thermometer(PRT)is installed on the blackbody cavity to monitor the temperature uniformity and stability of the blackbody radiation source and ensure its long-term operation reliability.Based on the above-mentioned importance of PRT in on-board calibration,an experimental study and calibration of the specially developed PRT's own characteristics have been carried out.Two calibration methods are used: the PRT is calibrated by the temperature reference point defined by the international temperature scale-phase change fixed point;The calibrated first-class standard platinum resistance thermometer is used as a standard device for comparison calibration in a thermostatic bath.The PRT temperature is traced back to the international temperature scale through calibration.The calibration results of the two methods are compared to ensure the validity of the calibration results.Obtaining the corresponding resistance value of PRT at each temperature point.Study on stability based on PRT.By selecting 14 PRTs with stable performance,the resistance value of the PRT at the water three-phase point is measured once a month.The annual stability is 0.003 ? after analyzing the accumulated experimental measurement data for one year.By comprehensively analyzing the uncertainty components of PRT measurement,the accuracy and effectiveness of the uncertainty evaluation results are ensured.Based on JJF1059.1-2012 "Evaluation and Expression of Uncertainty in Measurement"(GUM),the PRT's own characteristics are experimentally measured,including repeatability and self-heating effect,and thermal hysteresis,which has not been considered before,is comprehensively evaluated.For the uncertainty evaluation of 7temperature points uniformly distributed in the measurement temperature interval,taking the evaluation results of comparison method at 0?as an example,the uncertainty component results of experimental measurement:(1)the repeatability measurement of PRT at each temperature point is within 0.1 m K,and the standard uncertainty is 0.1 m K;(2)The thermal hysteresis of the measured PRT does not exceed1.1 m K in multiple temperature intervals,and the best estimation value is 0.55 m K,which is uniformly distributed;(3)The best estimate of the measured PRT at 0?is5.3 m K,and half of the best estimate is 2.65 m K in the variation interval.Combined with other uncertainty components obtained by searching relevant data,the standard uncertainty of high precision platinum resistance thermometer at 0?is 4.03 m K.At80?,the maximum measurement uncertainty is 4.9 m K.The uncertainty evaluation method of PRT based on Monte Carlo method(MCM)is proposed.The CVD equation is used as the measurement model of MCM.Run the Monte Carlo program in Matlab software to output the standard uncertainty of each temperature point.The program output standard uncertainty is 4.02 m K at 0 ? and 5.2 m K at 80 ?.Uncertainty data of seven temperature points obtained by the two evaluation methods are plotted as graphs,and it is found that the two methods are closely consistent,which verifies the reliability of PRT uncertainty evaluation results.