【作者】 原桂彬；

【导师】 戴景民； 孙晓刚；

【作者基本信息】 哈尔滨工业大学 ， 仪器科学与技术， 2007， 博士

【摘要】 导弹在助推段及飞行中段的红外辐射主要来自其羽焰,而其飞行的背景主要是云层。为了能有效地从云层背景中识别出导弹,必须研究导弹羽焰及云层背景的红外辐射特性。云层的红外辐射特性主要决定于云层中水滴粒子系和冰晶粒子系对太阳光的多次散射。本课题受航天支撑技术基金资助,旨在研究导弹羽焰及云层背景的红外辐射特性,以便为导弹预警的实时场景仿真提供红外辐射模型。论文的研究工作主要包括两部分:一、为了识别来袭导弹,研究了导弹羽焰真实温度及发射率的计算方法,提出了两种新的多光谱数据处理方法:基于亮温模型的逼近法和基于亮温模型的连续测量法。二、为了区分云层背景,计算了零距离时云层中水滴粒子系及冰晶粒子系的多次散射及云层发射率、透过率。为了计算远距离时云层的辐射亮度,提出了一种辐射传输方程的快速计算方法,即累积因子法。论文主要完成了以下几方面的研究工作:(1)提出了基于亮温模型的逼近法。目前大多数多光谱数据处理方法需要事先假设发射率与波长之间的函数关系,本文根据亮度温度与真实温度及波长的关系,当波长趋近于0时,亮度温度就是真实温度。由测量得到的各通道的亮度温度与对应的波长进行非线性最小二乘拟合,可以得到波长趋于0时的亮度温度,即为真实温度。对应大气的两个窗口3~5μm和8~12μm,分别在高温段(1500~300K)、中温段(500~1500K)和低温段(160~500K)进行了仿真试验。然后用实测数据对该算法进行了验证,并考察了基于亮温模型的逼近法对随机噪声的承受能力。仿真结果证明:基于亮温模型的逼近法可以识别出常见工程材料的真实温度,而且在低温段比在高温段有更高的计算精度。该方法对随机噪声的承受能力很弱。但可以把其计算结果作为其他多光谱数据处理方法的初值,避免初值估计的盲目性。(2)提出了基于亮温模型的连续测量法。在很窄的波段范围内及很短的一个连续测量周期内,假设光谱发射率在所选定的波长处与温度有近似的线性关系,通过处理两个不同时刻多光谱测温仪的测量数据,可以求得多个波长下的法向光谱发射率。仿真实验结果证明:基于亮温模型的连续测量法承受随机噪声的能力很强,无论有无噪声,该算法的计算结果都令人满意。利用该算法对实测数据进行处理,结果证明该算法的计算精度很高,能较好地解决目标真实温度的测量问题。(3)提出了基于超椭球方程计算任意形状粒子的等效半径的方法。球形粒子的散射计算要用Mie理论,而非球形粒子的散射计算要用T矩阵方法。T矩阵方法中首先要计算非球形粒子的等效半径,目前计算非球形粒子的等效半径的方法大都属于曲面积分方法,计算比较繁琐。本文利用空间超椭球方程的体积及表面积计算公式则可以很方便地计算出非球形粒子的体积等效半径和表面积等效半径,这样就可以对T矩阵方法进行改进,以提高计算速度。用FORTRAN语言编写了改进的T矩阵计算程序,并计算了零距离时卷云中非球形冰晶粒子系的散射特性,将计算结果与文献数据进行比对,证明对T矩阵方法的改进是成功的。论文还分析了粒子系的散射参数随等效半径的变化规律及修正的伽马分布的四个参数对粒子系散射特性的影响。(4)提出了累积因子法。为了计算远距离时云层的红外辐射特性,需要考虑辐射传输问题,求解辐射传输方程。目前辐射传输方程的求解方法如逐次迭代法、蒙特卡罗法等速度慢,不能用于在线实时仿真。本文提出的累积因子法,首先计算单次散射,然后求出累积因子,即多次散射对单次散射的比率,进而计算多次散射,即可计算出云层的辐射亮度。该算法可以快速有效地求解辐射传输方程,提高多次散射的计算速度。将累积因子法的计算结果与逐次迭代法和LOWTRAN的计算结果进行了比对。表明在相同计算精度下,累积因子法计算速度远比逐次迭代法快,可以用于在线实时仿真。更多还原

【Abstract】 The infrared radiation of missile during boosting phase and mid-course radiate mainly from its plume, and its background is most likely to be cloud. In order to distinguish attacking missile from cloud background effectively, it’s necessary to study the infrared radiation signature of missile plume and that of cloud background. The infrared signature of cloud depends on the multiple scattering of water droplets or ice crystals within clouds. This project is supported by foundation of spaceflight supporting technology. Its intention is to study the infrared signature of missile plume and that of cloud, to establish radiation models for real time scene simulation in missile early-warning system.This dissertation is composed of two parts. Firstly, in order to distinguish missile form cloud background, algorithm for true temperature of missile plume and its emissivity were studied. Two new data processing algorithms for multiple wavelength pyrometer were put forward. One algorithm is approach method based on brightness temperature (AMBT). The other one is continuous measuring method based on brightness temperature (CMBT). Secondly, in order to distinguish cloud background, the multiple scattering of water droplets and ice crystals in clouds were calculated. Emissivities and transmissivities of clouds were also computed. In order to calculate radiation brightness of cloud at a distance, a new algorithm to speedup the calculation of radiation transfer equation, build-up factor algorithm was proposed.Finished research works including:Ⅰ. Put forward a new multi-spectral data processing method, AMBT. In order to avoid guessing the relationship between emissivity and wavelength, according to the relationship among brightness temperature, true temperature and wavelength, it’s known that true temperature is equal to brightness temperature when the wavelength approaching to zero. Thus, the calculated true temperature can be derived through non-linear least squares fitting between brightness temperatures at different channels and the corresponding wavelengths. Simulation experiments were made at high temperature (1500~3000K), intermediate temperature (500~1500K) and at low temperature (160~500K) corresponding to the two atmospheric windows from 3 to 5 micron and from 8 to 12 micron. Actual measured data were used to validate the effectiveness of AMBT. Then stochastic noise was superimposed to evaluate its influence on the precision of AMBT. Simulation results prove that AMBT is more accurate used at low temperature than to be used at high temperature. This method can obtain true temperature of most engineering materials at a distance. Influence of stochastic noise on AMBT is uncertain. So we can use AMBT to estimate the preliminary value for other algorithms avoiding the blindness of estimation.Ⅱ. Described another new multi-spectral data processing method, CMBT. Supposing linear relation exists between emissivity and temperature at chosen wavelength during narrow waveband and in short measuring period, through processing measured data at two different time, spectral emmissivites at different wavelength can be obtained. Simulation results prove that CMBT can bear high random noise. All the calculation results are satisfied. Actual measured data was processed by CMBT and shows that CMBT is of high precision. This algorithm can be used in practical measurement and get better results.Ⅲ. Use superellipsoid equation to simulate nonspherical particles with arbitrary shapes and calculate their equivalent radius. Mie theory is used to calculate scattering of global particles, while T-matrix method is used to calculate scattering of nonspherical particles. Equivalent radius is very important in T-matrix method. Till now, most of the algorithms for equivalent radius of nonspherical particles are using integral on curved surface. They are inconvenient and time consuming. Using superellipsoid equation can shorten the calculating time for equal-volume-sphere radius and equal-surface-area-sphere radius. Program of the improved T-matrix method was compiled using FORTRAN language. Scattering parameters of ice crystals in cirrus was calculated. The computation results were compared with that of reference and proved that the improvement for T-matrix method is successful. Relations between scattering parameters and equivalent radius were analyzed. Modified Gamma distribution parameters’influence on the scattering signature of clouds was also researched.Ⅳ. A new algorithm of calculation of radiation transfer equation, build-up factor algorithm was proposed. In order to calculate infrared signature of clouds at a distance, radiation transfer must be taken into account. Algorithms of calculation of radiation transfer equation, such as successive iterative method and Monte Carlo method, are time-consuming and couldn’t be used online. As build-up factor algorithm, single scattering was calculated firstly. Then a build-up factor was defined to estimate the contribution of multiple scattering. The total scattering of particle system is the sum of single scattering and multiple scattering. Then the radiation brightness of cloud can be calculated. This algorithm can speedup the calculation of radiation transfer equation, calculate multiple scattering quickly. The results of this algorithm were compared with that of successive iterative method and that of LOWTRAN. Results show that to get the same calculating precision, build-up factor algorithm is much faster than successive iterative method and can be used in online simulation.更多还原