【作者】 孙崐；

【导师】 孙晓刚；

【作者基本信息】 哈尔滨工业大学 ， 仪器科学与技术， 2012， 博士

【摘要】 爆炸火焰温度是表征导弹、炸药爆炸威力的重要参数，对其测量技术开展研究对于国防、化学、材料等领域有着十分重要的现实意义。由于爆炸过程具有破坏性及瞬时性，使得对爆炸火焰温度进行测量具有一定的难度。本课题受国家级项目资助，目的是对爆炸火焰真温测量技术进行研究，为评估爆炸威力提供依据。论文紧紧围绕爆炸火焰真温测量这一主题，对国内外测量爆炸火焰温度已有的方法和仪器进行研究，针对其不足加以创新与改进，成功研制可用于测量爆炸火焰真温的多光谱高温计；针对目前高温计标定领域存在的问题，提出了两种无源温区标定方法：基于幂函数的无源温区标定方法和基于对数函数的无源温区标定方法，两种标定方法的提出对于完善辐射测温理论、拓宽辐射高温计的测温范围、提高无源温区测量精度具有重要的意义。目前大多数辐射高温计存在两方面的问题：一、在未知发射率的条件下，只能测量物体的亮温而非真温，采用估计发射率的方法误差较大；二、防爆性差、响应速度慢、采样率低及测量范围小。针对以上问题，本文研制了测量爆炸火焰真温的多光谱高温计。以辐射测温理论为基础，采用二次测量法和亮温逼近法相结合解决真温测量理论问题；提出了光学远传结构使高温计的主体与光学瞄准系统分离，提高了高温计的防爆性能；设计了自动选择量程功能拓宽了高温计的测量范围；设计了自启动功能大大增强了高温计的自动化程度；设计了超高速同步数据采集系统增加了高温计的响应速度与采样率；加入了无线传输与上位机的结构，保证了工作人员的安全。为了解决目前无源温区标定方法存在的问题，提出了基于幂函数模型的无源温区标定方法。利用幂函数曲线图形与高温计有源温区的温度—电压曲线具有很高相似性，建立以幂函数为基础的数学模型，应用对曲线走势有很强控制能力的导数最小二乘法对模型求解，从而实现对无源温区的标定。在此基础之上，应用Plank定律验证此标定方法的理论精度，同时采用有源温区的实测数据精度验证与无源温区的理论验证共同实现对此方法在实际应用中精度的验证。实验结果证明：此方法外推温度区间3100℃~3500℃，实测数据精度优于0.7%。提出了基于对数函数模型的无源温区标定方法。借助于高温计的传递函数建立了温度—电压曲线，并应用导数最小二乘法求解模型。分别应用黑体辐射出度及高温计有源温区的实际标定数据对此方法进行了精度验证。实验结果证明：此标定方法在理论上具有极高精度，在实际应用中的精度令人满意，外推温度区间3100℃~3500℃，实测数据精度为1.1%。为了确定两种标定方法的使用条件进行了大量的仿真实验。通过分析不同波长和不同温度点个数对两种方法标定精度的影响分别确定了两种方法的应用点数与波长；通过分析不等间隔温度点及温度点选取区间对两种方法精度的影响，得到了温度点选取的理论依据；通过对两种方法进行了抗随机误差能力的分析，得到两种方法适用的精度范围。仿真实验结果证明：基于幂函数模型的标定方法使用条件容易实现，且抗随机误差能力强；基于对数函数模型的标定方法对使用条件要求较为苛刻，抗随机误差能力弱。利用所研制的多光谱高温计分别进行了黑体炉模拟实验及TNT炸药爆炸实验。通过快速遮挡高温黑体输出窗口仿真快速变化的温度场以考察此高温计对于快速变化的温度场的响应，并在野外对3kgTNT炸药爆炸进行了测量，分析了实验结果并给出了此高温计的测量不确定度。在全量程范围内，此多光谱高温计的测量不确定度为1.2%。更多还原

【Abstract】 Explosion flame temperature is an important parameter that characterizesthe explosive forces of missiles and explosives. Its research has practicalsignificance in the field of national defense, chemistry, and material sciences etc.Because of the destructiveness and instantaneity in the explosion process, it israther difficult to measure the explosion flame temperature. This research,sponsored by funds on the national level, researched on the true temperaturemeasurement technology of the explosive flame so as to lay foundation for theevaluation of explosive force.The existing methods and instruments in the measurement of the explosiveflame true temperature were researched on and improvement was made directingat the existing faults. The multi-spectral pyrometer was successfully developedfor this purpose. For the high temperature calibration, two non-sourcetemperature calibration methods were proposed, one being power-function based,and the other being logarithmic function based. The two methods proposed aresignificant in improving the radiation temperature measurement theory,broadening the measurement range of the pyrometers, and upgrading themeasurement precision of non-source temperature.At present, most radiation pyrometers have the following two faults. First,they can only measure the brightness temperature rather than the true temperaturewhen the emissivity is unknown. The emissivity estimation method usuallyresults in comparatively large error. Secondly, it has low anti-explosion ability,low response speed, low sample rate and narrow measurement range. To solvethe problem mentioned above, a multi-spectral pyrometer was developed tomeasure the true temperature of the explosive flame in this research. Based onthe radiation temperature measurement theory, the Second Measurement Methodand Brightness Temperature Approximation Method were combined to resolvethe theoretical issue in the true temperature measurement. An optical structurewas established to separate the main body of the pyrometer and the opticaltargeting system, which improved the anti-explosion property of the pyrometer.The automatic selection of the measurement range was designed to widen the range of the pyrometer. Automatic activation designed greatly increased theextent of automation of the pyrometer. high-speed synchronized data collectionsystem was developed which improved the response rate and sampling rate. Thewireless transmission and host computer were implemented to ensure the safetyof the personnel.In order to solve the existing problem in the non-source temperaturecalibration, a non-source temperature calibration method was proposed based onthe power function. Since the curve of power function and that of thetemperature-voltage in source temperature region have great similarity, amathematical model was established based on the power function. Since theDerivative Least-Square has high controllability over the shape of the curve, itwas used to obtain the solution of the model so as to realize the non-sourcetemperature calibration. In addition, Plank Law was used to testify the theoreticalprecision of this calibration method. The experimental data collected in thesource temperature range and the theoretical testing in the non-sourcetemperature range were employed to testify the precision of this method in thepractical application. The test results show that the extrapolation temperaturerange is between3100℃and3500℃,and the actual experimental data precisionis greater than0.7%.A calibration method in non-source temperature range was proposed basedon the logarithmic function model. The transfer function of the pyrometer wasused to establish the temperature-voltage curve and the solution to the model wasobtained using derivative Least-Square. The blackbody emissivity and the actualcalibration data of the pyrometer in the source temperature range were usedrespectively to testify the precision of this method. The result shows that thecalibration has both high theoretical precision and satisfactory precision inpractical applications. The extrapolation temperature range is between3100℃and3500℃, and the actual data measurement precision is1.1%.In order to define the applicable condition of the two calibration methods,large amount of simulations were conducted. Through the analyzing the influenceof different wavelengths and the numbers of temperature points on the precisionof the two methods, the wavelength and number of points were determined forthe two methods respectively. Through the analysis of the influence of temperature points at unequal intervals and the range of temperature selection onthe precision of the two methods, the theoretical basis of temperature pointselection was established. Through the analysis of anti-random-error ability ofthe two methods, their applicable precision ranges were determined. Thesimulation results showed that the conditions for the calibration method based onthe power function model could be easily realized and the ability of anti-random-error was strong, while the method based on the logarithmic function modelrequired strict application condition and the ability of anti-random-error wasweak.The blackbody furnace simulation and TNT explosion experiment wereconducted using multi-spectral pyrometer developed in the research. The reactionof this pyrometer to temperature field was tested through rapidly shielding thehigh-temperature blackbody output window to simulate the rapidly changingtemperature field. In the field experiment, the3kg TNT explosive was tested andthe experimental results were analyzed. The uncertainty of the measurement ofthe pyrometer is obtained which is1.2%in the overall measurement range.更多还原