【作者】 徐艳春；

【导师】 杨春玲；

【作者基本信息】 哈尔滨工业大学 ， 电力电子与电力传动， 2010， 博士

【摘要】 在红外物理领域,发射率是材料重要的红外辐射特性参数,它的有效测量在工农业和国防领域有着重要的意义。然而物理上的普朗克定律因数学欠定问题无法从辐射信号中分离出目标的发射率和真温。近年来发展起来的多光谱法是一种有效的解决方法,但是需要对辐射光电信号按波长进行分光,波长数越多多光谱处理结果精度越高,但每个工作波段内的辐射能量就变得非常微弱,这些信号淹没在强噪声背景下,传统的光电信号检测方法已不能满足此项技术的需要,因此必须研究强噪声背景下微弱光电信号检测技术。此外,我国目前正在研制高光谱卫星,波长需分辨到纳米级,也需要研究更高精度的微弱光电信号检测技术。在这些高精度的微弱光电信号检测系统中,为了减小零点漂移对恒定微弱光电信号的影响,首先通过调制盘将光电信号调制成周期信号,从而在传感器上接收到含噪声的微弱周期信号,以进一步实现检测。研究表明,混沌理论对任何零均值噪声均具有极强的“免疫力”,而对微弱的周期信号却很敏感,因此可以很好地解决目前微弱光电信号检测存在的问题。首先分析了典型混沌系统的数学模型及动态特性,深入研究了微弱光电信号幅值和频率对典型混沌系统动态特性的影响,利用典型混沌系统对信号幅值和频率的敏感性及系统相变情况,确立了微弱光电信号检测系统中的最优混沌模型。在最优混沌模型的基础上,进一步对最优混沌模型动态特性进行分析,提出将李亚普诺夫指数法引入到最优混沌模型中,从定量的角度计算混沌系统的阈值,并应用周期系数微分方程理论和小数据量法进行微弱光电信号幅值的检测,从而确立了微弱光电信号检测系统中最优混沌模型可检测的最低信噪比。为了检测微弱光电信号的未知频率,提出了基于比例微分控制策略的R(o|¨)ssler混沌系统检测微弱光电信号频率的方法,通过比例微分控制策略实现对R(o|¨)ssler混沌系统的周期控制,然后利用谱分析的方法,实现任意位置的信号频率检测。在此基础上,针对R(o|¨)ssler混沌系统阶数较高,控制过程复杂的问题,提出了基于混沌模型的自适应频率检测方法,通过对待检信号频率与Duffing系统周期策动力频率间的频差控制,实现频率检测中的混沌模型自适应选择,从而实现检测强噪声背景下微弱光电信号频率的目的。最后,在基于最优混沌模型的微弱光电信号幅值和频率检测方法的基础上,建立了基于混沌振子的微弱光电信号检测实验系统,并对微弱光电信号的幅值和频率检测进行了实验研究。实验结果验证了基于混沌振子的微弱光电信号检测方法的正确性和有效性。同时,将该检测系统应用于双向反射分布函数测量系统中,实现了双向反射分布函数测量中微弱光电信号的检测。更多还原

【Abstract】 At the infrared physics field, emissivity is an important property parameter of the materials. There is a significant meaning to measure it. However because there is a mathematic problem not to be solved very well in physics field, the emissivity and true temperature of the target can’t be separated from radiation signal. In recent years multi-spectrum method is developed to solve the problem. But it is needed to split the radiation photo-electric signal according to wave length. When there are a lot of the splitted wavelength numbers, the precision is higher in processing results. However the radiation energy is very weak in every working waveband. These signals are submerged in strong noise. The traditional detection method can’t meet the requirements. So the weak signal detection technology in strong noise background must be researched. Moreover high spectral satellite is being researched in our country. It also needs to study on weak photo-electric signal detection.In the high precision detection system of weak photo-electric signal, chopper weel is used in order to decrease zero drift influence to constant weak photo-electric signal. At first, the weak photo-electric signal is modulated periodic signal. Then the signal is received in sensor. Of course the signal is submerged in strong noise.The results show that chaos system has a very strong immunity to noise but it is sensitive to the periodic signal. So chaos theory can solve the difficult problem very well in weak photo-electric signal detection system.The mathematic model and dynamic property of typical chaos system is analyzed firstly and the influence of the amplitude and frequency of weak photo-electric signal to chaos dynamic property is researched. It is known that the typical chaos system is sensitive to the amplitude and the frequency of weak photo-electric signal. So the optimal chaos model is established in the detection system of weak photo-electric signal.On the basis of optimal chaos model, the dynamic property of optimal chaos model is analyzed deeply. And the method is proposed that lyapunov exponent method is introduced to optimal chaos model to calculate the chaos threshold quantitatively. So the amplitude of weak photo-electric signal is detected precisely. Then the lowest signal-noise ratio in weak photo-electric signal detection system can be determined.To detect the unknown frequency of weak photo-electric signal, the frequency detection method based on proportional differential control of R(o|¨)ssler chaos system is proposed. It can detect unknown frequency through controlling the chaos state to big periodic state of R(o|¨)ssler system firstly and using spectrum analysis. On the basis of the method, adaptive frequency detection method based on chaos model is proposed. It can solve the problem of the order number high of R(o|¨)ssler system and complex control. The frequency diviation between the frequency of Duffing system and the detected frequency is controlled to choose the chaos model adaptively. Thus the frequency of weak photo-eletric signal can be detected accurately.Finally, on the basis of the amplitude and the frequency detection methods of weak photo-electric signal based on optimal chaos model the experiment system of weak photo-electric signal is developed. And the the amplitude and the frequency of weak photo-electric signal is detected in the experiment. The experiment results verified that it is right and effective to the detection method of weak photo-electric signal based on chaos oscillator. Meanwhile the detection system based on chaos oscillator is used in bidirectional reflectance distribution function measurement system and the weak photo-electric signal in the bidirectional reflectance distribution function system is detected precisely.更多还原