典型振动信号时-频分析方法研究

首先对比较简单的单分量信号和多分量信号进行时-频分析,对比了不同时-频分析方法(STFT、WD、Hilbert时-频谱分析、连续小波分析)的分析精度。结果表明对单分量信号和可以分解为若干单分量信号叠加的多分量信号,Hilbert时-频谱分析精度最好,WD和STFT次之,连续小波变换的精度最差;由于Hilbert时-频谱分析和WD方法要求信号为单分量信号或可以分解为若干单分量的信号,故其不适用于复杂信号的时-频分析;对多分量信号,STFT时-频谱较连续小波时-频谱精度高些。需要指出的是,当使用STFT方法时,取不同的窗参数,STFT的时-频分析精度相差很大,必须使用合适的窗参数,以取得较好的分析效果。此外还对El Centro波的两水平分量和人工合成的地震动进行了时-频分析,虽然STFT的时间分辨率和频率分辨率存在矛盾,但它对信号的适应性最强,对复杂信号的时-频分析精度一般可以满足需要,而连续小波变换进行信号的时-频分析时,时-频分辨精度最差,且对于大于30Hz的频率分量,随频率的增加,时-频分析的效果和精度迅速变差,说明小波变换并不适合于做时-频分析,尤其对于高频信号,用连续小波变换得到的时-频谱图看起来很粗糙,而且很难准确解释。进一步对比了真实地震动和人工地震动的时-频分布,认为用胡聿贤教授合成人工地震动的方法合成的地震动在时-频域上的分布和真实地震动相似。     

Comparative study of As,Cd, Cu,Cr, Mg,Mn, Ni,Pb and Zn concentrations between sediment and water from estuary and port

The contents of As, Cd, Cu, Cr, Mg, Mn, Ni, Pb and Zn have been determined in sediment and water samples from Valle de las Garzas estuary and Port Manzanillo (Colima, Mexico) using ICP-AES. The concentrations of these elements were used for a comparative study to determine the distribution of heavy metals and to evaluate which elements reflect natural or anthropogenic backgrounds. For this purpose, seven sampling points were selected: Four of them correspond to the lagoon, and three were situated in the port. Statistical analysis of the mineral content was assessed. Initially, data comparison was assessed by statistical tests for each variable. Principal component analysis was then applied considering the influence of all variables at the same time by obtaining the distribution of samples according to their scores in the principal component space. In this way, four studies were carried out: (1) study of sediments collected during the dry season; (2) study of sediments collected during the rainy season; (3) comparative study between sediments from rainy and dry season; and (4) study of water composition collected during rainy season. From the results of the performed analyses, it can be concluded that metals distribution pattern reflected natural and anthropogenic backgrounds (e.g., sediments from the lagoon, situated at the beginning of the rain channel, presented high contents of Zn and Cu, perhaps related to anthropogenic activities or the influence of igneous sediments).     

高分辨率影像城市植被自动提取算法

近十几年来,随着城市化进程加剧,准确获取城市植被的分布信息,是城市气候和地表能量平衡研究的重要内容。高空间分辨率遥感影像数据,为精确获取和动态监测城市植被提供了重要资料。本研究利用资源三号数据对长江三角洲地区城市植被进行光谱特征分析与提取,提出一种城市植被的自动化信息提取算法—分离面法（Hyperplanes for Plant Extraction Methodology,HPEM）。结果表明：在假彩色反射率空间,植被与NDVI值低的背景有很好的分离性,而在真彩色反射率空间,植被与NDVI值高的背景有很好的分离性;HPEM能很好地避免NDVI最佳阈值法中将建筑物误分为植被的问题,其精度明显优于NDVI最佳阈值法,Kappa系数从0.85提高到0.90,总的错分与漏分误差从21.15%降低到14.18%。可见,本文的HPEM方法能有效提高城市植被信息自动提取的精度。     

Ar-Ar laser dating of tektites from the Cheb Basin (Czech Republic): Evidence for coevality with moldavites and influence of the dating standard on the age of the Ries impact

Moldavites (Central European tektites) are genetically related to the impact event that produced the ∼24-km diameter Ries crater in Germany, representing one of the youngest large impact structures on Earth. Although several geochronological studies have been completed, there is still no agreement among ⁴⁰Ar-³⁹Ar ages on both moldavites and glasses from Ries suevites. Even recently published data yielded within-sample mean ages with a nominal spread of more than 0.6 Ma (14.24-14.88 Ma). This age spread, which significantly exceeds current internal errors, must be in part ascribed to geological and/or analytical causes.This study reports the results of a detailed geochronological investigation of moldavites from the Cheb area (Czech Republic), which have never been dated before, and, for comparison, of two samples from type localities, one in southern Bohemia and the other in western Moravia. We used ⁴⁰Ar-³⁹Ar laser step-heating and total fusion techniques in conjunction with microscale petrographic and chemical characterization. In addition, with the purpose of ascertaining the influence of the dating standards on the age of the Ries impact and making data from this study and literature consistent with the now widely used Fish Canyon sanidine (FCs) standard, we performed a direct calibration of multi-grain splits of the Fish Canyon biotite (FCT-3) with FCs. The intercalibration factors (), determined for eight stack positions in one of the three performed irradiations, were indistinguishable within errors and gave an arithmetic mean and a standard deviation of 1.0086 ± 0.0031 (±2σ), in agreement with previous works suggesting that biotite from the Fish Canyon Tuff is somewhat older (∼0.8%) than the coexisting sanidine.Laser total fusion analysis of milligram to sub-milligram splits of five tektite samples from the Cheb area yielded mostly concordant intrasample ⁴⁰Ar-³⁹Ar ages, and within-sample weighted mean ages of 14.66 ± 0.08-14.75 ± 0.12 Ma (±2σ internal errors, ages relative to FCs) that overlap within errors. These ages match those obtained for samples from western Moravia (14.66 ± 0.08 Ma) and southern Bohemia (14.68 ± 0.11 Ma), supporting the genetic link between Cheb Basin tektites and moldavites, and, consequently, between Cheb Basin tektites and the Ries impact. In contrast to samples from the Cheb area and Moravia, ⁴⁰Ar-³⁹Ar ages from total fusion experiments on the Bohemian specimen ranged widely from ∼14.6 to ∼17.0 Ma. Older apparent ages, however, were systematically obtained from fragments characterized by visible surface alteration. Laser step-heating experiments, although displaying slightly disturbed age profiles, were in line with total fusion analyses and yielded well-defined plateau ages of 14.64 ± 0.11-14.71 ± 0.11 Ma (±2σ internal errors, ages relative to FCs).A thorough comparison of our and previous ⁴⁰Ar-³⁹Ar ages on both moldavites and Ries suevite glasses, recalculated relative to the ⁴⁰Ar^∗/⁴⁰K ratio recently determined for FCs using intercalibration factors available in or derivable from the literature, reveals some inconsistencies which may be ascribed to either geological or analytical causes. Based on our data, decay constants in current use in geochronology, and ages calculated relative to FCs, we infer that the age of moldavites is 14.68 ± 0.11 Ma (±2σ, neglecting uncertainties in the ⁴⁰K decay constants).     

淮河源区中更新世黄土堆积的元素地球化学特征及其古气候意义

利用地球化学元素分析方法,对淮河源区中更新世黄土、古土壤剖面含有的化学元素及多种化学元素指标进行了分析。研究表明:区内中更新世以来的气候变化,主要以湿热气候为主,淋溶及氧化作用较强;表现在气温上HT-1~HT-3的气温较低,HT-4~HT-7的气温高,HT-8的气温又相对较低,其变化趋势表现为中更新世经历了从早期寒冷转湿热转温干,中期凉干转暖湿到晚期温湿转暖湿。     

Enhanced resistance of Portunus trituberculatus to Vibrio alginolyticus by selective breeding

We established a line (screened) of Portunus trituberculatus by selectively breeding individuals that survived from challenge with Vibrio alginolyticus, and compared the response of screened and unscreened (control) P. trituberculatus challenged with V. alginolyticus. We measured superoxide dismutase, catalase, acid phosphatase, alkaline phosphatase, and peroxidase activity and the content of hemocyanin in the plasma and phenoloxidase activity in serum. The cumulative survival rate after 24-h challenge with V. alginolyticus was significantly higher in the screened crabs than in the unscreened crabs (P<0.05). T-SOD and PO activity were significantly lower in the screened stock than in the unscreened stock (P<0.05). POD, CAT, and ACP activity and hemocyanin content were significantly higher in the screened stock than in the unscreened stock. Our results suggest that the screened stock was more resistant to infection. Furthermore, the indices we measured may be used to evaluate the health state of P. trituberculatus.     

加强目标观测,服务防灾减灾

以深入浅出的方式介绍了目标观测的定义、历史、确定目标观测敏感区的方法,以及实施目标观测与防灾减灾的关系;并以欧洲中期天气预报中心目标观测业务化为例,简明地概括了目标观测的实施过程;结合国内目标观测研究现状,对中国目标观测的未来提出了看法。     

基于遗传算法的聚焦爬虫搜索策略设计与研究

网络爬虫是搜索引擎的重要组成部分。针对目前聚焦爬虫搜索策略的不足,提出了一种新的搜索策略解决方案。在搜索过程中对适应度高于或低于种群平均适应度的个体采用不同的交叉概率和变异概率来扩大爬虫的爬取范围、增加新个体,并通过改进遗传算子,提高聚焦爬虫的搜索效率。实验证明,基于自适应遗传算法的聚焦爬虫在一定程度上解决了传统遗传算法的"早熟"问题,而且能够爬取到更多主题相关的网页和相关度高的网页。     

二层交换机转发帧的原理与实现

介绍了二层交换机转发帧的原理及其具体的硬件和软件实现.着重讲解了其软件实现的过程中,在各种端口上设置MAC地址的处理方法.     

基于密度的线数据分组算法研究

目前,地理空间数据面临着由于数据量膨胀和计算量高速增长而引起算法效率低的问题,采用"分而治之"的数据分组策略提高运算效率已成为研究的热点。面向分布不均匀的线数据,本文提出了基于密度的线数据分组算法(简称LGAD)。首先,算法通过查找高密度区提取样本线段,保证了分组算法的起点落到高密区;其次,考虑线空间拓扑关系的复杂性,引用水平、垂直和夹角距离度量线段间距离,创建样本线段与其他线段的距离矩阵;最后,以距离矩阵和最优选择方法实现数据负载均衡分组。实验结果显示,对数据分组和分组后数据进行线段聚类的2个过程中,该算法体现了较好的时间优势,与串行计算相比,在分组数为2-12 时,平均比率达4.3,提高了应用的响应速度,具有较好的实际意义。