基于DOAS技术的气溶胶粒谱分布反演方法研究

大气气溶胶不仅对全球的气候变化产生重大影响,其本身也是一种污染物,另外它在许多污染气体的化学反应中起重要作用。因此,实时监测大气气溶胶已成为环境领域的重要研究方向。差分吸收光谱技术是一种基于痕量气体“指纹”特性反演其浓度的光学遥感方法,同时该方法也可用于大气气溶胶消光系数的测量。文章介绍了利用闪烁差分吸收光谱系统监测大气气溶胶粒谱分布的方法,重点阐述了基于蒙特卡罗方法的粒谱分布反演算法,监测结果通过与PM10、能见度及Angstrom波长指数的对比证实了该方法的可行性,为近地面大气气溶胶监测提供了新的手段,同时也扩展了差分吸收光谱技术的应用范围,该方法对大气化学的研究有着重要的意义。     

Using scatterplots to understand and improve probabilistic models for text categorization and retrieval

The two-dimensional representation of documents which allows documents to be represented in a two-dimensional Cartesian plane has proved to be a valid visualization tool for Automated Text Categorization (ATC) for understanding the relationships between categories of textual documents, and to help users to visually audit the classifier and identify suspicious training data. This paper analyzes a specific use of this visualization approach in the case of the Naive Bayes (NB) model for text classification and the Binary Independence Model (BIM) for text retrieval. For text categorization, a reformulation of the equation for the decision of classification has to be written in such a way that each coordinate of a document is the sum of two addends: a variable component P(d|c_i), and a constant component P(c_i), the prior of the category. When plotted in the Cartesian plane according to this formulation, the documents that are constantly shifted along the x-axis and the y-axis can be seen. This effect of shifting is more or less evident according to which NB model, Bernoulli or multinomial, is chosen. For text retrieval, the same reformulation can be applied in the case of the BIM model. The visualization helps to understand the decisions that are taken to order the documents, in particular in the case of relevance feedback.     

A survey of literature on automated storage and retrieval systems

Automated Storage and Retrieval Systems (AS/RSs) are warehousing systems that are used for the storage and retrieval of products in both distribution and production environments. This paper provides an overview of literature from the past 30 years. A comprehensive explanation of the current state of the art in AS/RS design is provided for a range of issues such as system configuration, travel time estimation, storage assignment, dwell-point location, and request sequencing. The majority of the reviewed models and solution methods are applicable to static scheduling and design problems only. Requirements for AS/RSs are, however, increasingly of a more dynamic nature for which new models will need to be developed to overcome large computation times and finite planning horizons, and to improve system performance. Several other avenues for future research in the design and control of AS/RSs are also specified.     

Operational retrieval algorithms for JEM/SMILES level 2 data processing system

To measure the thermal emission from stratospheric minor species with high sensitivity, the Superconducting Submillimeter-wave Limb-Emission Sounder (SMILES) aboard the Japanese Experiment Module (JEM) of the International Space Station (ISS) carries 4 K cooled Superconductor–Insulator–Superconductor (SIS) mixers. The major feature of the SMILES is its high-sensitive measurement ability with low system noise temperature less than 700 K.As a part of the ground system for the SMILES, a level 2 data processing system (DPS-L2) has been developed. It retrieves the density distributions of the target species from calibrated spectra in near-real-time. The retrieval process consists of two parts: the forward model, which computes radiative transfer, and the inverse model, which deduces atmospheric states. Since the forward model must provide the most accurate basis for results and be implemented under limited computing resources, the forward model algorithm for an operational system has to be accurate and fast. Hence, the algorithm is improved (1) by designing accurate instrument functions such as the instrumental field of view (FOV), sideband rejection ratio of sideband separator, and spectral responses of acousto-optic spectrometer (AOS) and (2) by optimizing radiative transfer calculation.This paper presents the development of the DPS-L2 along with the details on its algorithm and the algorithm performance. The accuracy of this algorithm is better than 1%, and the processing time for single-scan spectra is less than 1 min with eight parallel processings using a 3.16-GHz Quad-Core Intel Xeon processor. Thus, this algorithm is suitable for the SMILES measurement.     

Channel selection using information content analysis: A case study of CO2 retrieval from near infrared measurements

A major challenge in retrieving CO₂ concentrations from thermal infrared remote sensing comes from the fact that measurements in the 4.3 and 15 μm absorption bands (AIRS or TES) are sensitive to both temperature and CO₂ variations. This complicates the selection of absorption channels with maximum CO₂ concentration information content. In contrast, retrievals using near infrared (NIR) CO₂ absorption bands are relatively insensitive to temperature and are most sensitive to changes of CO₂ near the surface, where the sources and sinks are located. The Orbiting Carbon Observatory (OCO) was built to measure reflected sunlight in three NIR spectral regions (the 0.76 μm O₂ A-band and two CO₂ bands at 1.61 and 2.06 μm). In an effort to significantly increase the speed of accurate CO₂ retrieval algorithms for OCO, we performed an information content analysis to identify the 20 best channels from each CO₂ spectral region to use in OCO retrievals. Retrievals using these 40 channels provide as much as 75% of the total CO₂ information content compared to retrievals using all 1016 channels in each spectral region. The CO₂ retrievals using our selected channels have a precision better than 0.1 ppm. This technique can be applied to the retrieval of other geophysical variables (e.g., temperature or CH₄), or modified for other instruments, such as AIRS or TES.     

Pressure broadening coefficients of induced by for Venus atmosphere

Carbon dioxide (CO₂) induced pressure broadening coefficients of water vapor (H₂O) lines have been determined using a terahertz time-domain spectrometer (TDS). Thirty-two rotational transitions of H₂O were observed in the spectral range of 18– (550–3050 GHz) for the first time. Using TDS allows one to measure absorption spectra with one order of magnitude better precision than Fourier transform spectrometer in this frequency region. The precision of our broadening coefficient measurements was 2.4% in average. The measured CO₂ induced pressure broadening coefficients are compared to those calculated by the complex Robert–Bonamy formalism. The difference between the measurement and the theoretical estimation was in the range of -10.7% to +19.0% confirming the credibility of the theoretical approach. The impact on retrieval of water vapor abundance was examined by performing inversion analysis on H₂O spectra of Venus atmosphere obtained with the Submillimeter Wave Astronomy Satellite. In this example case, the retrieved water vapor mixing ratio reduces by half at the altitude region of 70–85 km when applying the newly measured broadening coefficient compared to the air-broadening coefficient, and changes by 5% compared to that estimated by the complex Robert–Bonamy formalism.     

The impact of Martian aerosols on the retrieval of temperature profiles from PFS measurements

Our purpose was a qualitative assessment of the impact of dust and water ice aerosols on the retrieved temperature profiles and the retrieval process itself in the Martian atmosphere. It aims to quantify the related uncertainties in the atmospheric temperature profiles derived from radiance measurements of the Planetary Fourier Spectrometer (PFS), currently operating on the Mars Express orbiter. In this study the effects of aerosol opacities on simulated data and retrieved temperature profiles were also investigated.From the analysis of the model atmosphere including dust and water ice with different size distributions it results that the dust component affects weighting functions and brightness temperatures less than water ice. A similar situation is also observed when different vertical distributions are considered. Unlike dust, water ice with different sizes of crystals evidently influences weighting functions and brightness temperatures. The impact of the considered water ice vertical distributions on brightness temperatures is noticeable only near 840 cm⁻¹.Considering different dust opacities, the largest differences—5 K maximum—between retrieved temperature profiles were observed close to the surface, regardless assumptions on a size distribution or the refractive indices. Contrary to dust, the different sizes of water ice particles assumed during retrieval stronger affected the retrieved temperature profiles than water ice opacities. Moreover, the differences in the retrieved temperature profiles were amplified while wrong optical properties for dust as well as for water ice aerosol were assumed instead of the nominal case. This means that the wrong assumption can induce an additional source of the retrieval error and lead to unreasonable temperature profiles. In the cases of expected heavily loads water ice crystals, their size distribution in the Martian atmosphere should be known from other observations before the retrieval of the temperature profile is attempted.For the analyzed examples of real PFS measurements the impact of different dust vertical distributions on the retrieval of temperature profile is prominent only in layers close to the surface. However, these differences remain comparable with retrieval errors. All influences of dust on weighting functions, brightness temperatures and during retrieval can be neglected if the noise equivalent radiance (NER) of PFS is taken into account.     

双光路DOAS系统获取大气颗粒物粒谱分布方法研究

当前大气复合污染日趋严重,造成大气氧化性增强,气体向颗粒物的转化加快。大气颗粒物粒径大小及谱分布决定其在大气中的行为,以差分吸收光谱法（DOAS）为基础,结合双光路设计技术,开展实时、在线、获取近地面大气气溶胶颗粒物的粒谱分布的光谱方法研究。首先构建低噪声性能稳定的宽带氙弧灯为光源的双光路差分吸收光谱系统,基于干净天气条件下大气的能见度数据对系统进行校准,通过两个不同光路获得的光谱信号强度之比获取近地面紫外-近红外波段的大气总宽波段消光系数。基于宽波段消光系数,在去除瑞利散射以及气体吸收对消光系数的影响后,解析出气溶胶颗粒物的消光系数。基于核函数准则,利用均匀球型粒子的电磁场Mie理论来反演气溶胶物理特性,获得气溶胶粒子在该测量谱段的体积谱分布,利用体积谱与数密度谱的关系,反演出气溶胶粒子的数密度谱分布。开展利用直方图方法来表现颗粒物的粒谱分布方法研究,首先将DOAS测量波段近似等分为若干谱段,利用谱段处平均值,获取气溶胶粒谱直方分布图。最后把该系统和方法应用于外场实验,获得了气溶胶颗粒物在300～650 nm范围内的消光系数,将测量波段等分为11个谱段,反演了颗粒物的在0.1～1.25 μm粒径范围的数密度谱分布。该研究为整治我国灰霾天气,研究大气气相/粒子非均相化学反应提供科学依据。同时将推动DOAS技术的进一步发展和应用。     

机器学习的IBBCEAS光谱反演波段优化

非相干宽带腔增强吸收光谱技术（IBBCEAS）利用高精密谐振腔增强吸收光程,实现对痕量气体的高灵敏探测。目前,IBBCEAS技术主要采用发光二极管（LED）作为非相干光源。当谐振腔镜片反射率曲线与带宽有限的LED辐射谱不能很好匹配时,光谱反演波段选择不当可能会对被测气体浓度拟合结果产生较大偏差。以定量探测大气NO₂浓度为例,分析了IBBCEAS光谱反演波段对NO₂拟合结果的影响,发现当反演波段宽度窄到一定程度后,NO₂浓度拟合相对误差会迅速增加。为此,提出了一种基于RBF神经网络结合遗传算法的机器学习IBBCEAS光谱反演波段优化方法,以使浓度拟合误差达到最小。在430～480 nm待选波段内,选择各种宽度和中心波长的子波段作为反演波段,分别进行NO₂浓度拟合,以此获得435个样本数据,并将样本数据按照4∶1比例分成学习样本和测试样本,分别用于RBF神经网络学习训练和测试,得到输入参数“反演波段的起始波长与截止波长”与输出参数“浓度拟合相对误差”之间的非线性映射关系。使用遗传算法搜索最优反演波段,将反演波段的起始波长和截止波长组合进行个体编码,随机产生若干个体形成种群。以RBF神经网络的输出（即浓度拟合相对误差）作为个体适应度,经过多代种群进化过程后,获得适应度最优个体,即获得最优反演波段。在种群规模为100个体,种群进化最大代数为100的情况下,当种群进化第61代时,最优个体出现,对应的最优适应度为3.584%,最优反演波段为445.78～479.44 nm。选择相同带宽的其他4个典型反演波段,与最优反演波段下的NO₂拟合结果进行了对比。结果显示,在最优反演波段下,无论是拟合误差、相对拟合误差还是拟合残差标准偏差,均低于其他4个反演波段,光谱拟合质量达到最优。结果表明,利用机器学习来确定IBBCEAS最优反演波段是可行的。     

信息熵在粗糙集信息检索模型中的应用

在信息检索过程中,由于文档中存在大量的多义和近义现象,导致不确定性出现,这将影响检索的性能.为此我们采用信息熵和粗糙集理论来处理这类不确定性问题.首先计算训练文档集中的词之间的信息熵,对信息熵做模糊聚类来构造词之间的等价关系,然后借助于该等价关系提出并实现了一个以粗糙集上下近似为基础的信息检索模型,通过实验的测试,该模型能够提高信息检索的效率.