复指数方法降噪技术及其试验研究

实测信号中的噪声,以及模型阶次的不确定性给模态参数的准确识别带来困难。以提高模态参数识别精度为目标,提出基于模型定阶和信噪分离的复指数模态参数识别方法。该方法借助奇异值分解技术确定模型阶次,采用结构低秩逼近方法进行信噪分离。在此基础上,利用复指数法进行模态参数识别。分别选取一维的悬臂梁模型和二维的悬挂板模型进行物理模型实验,结果表明:该方法提高模态参数的识别精度,尤其是阻尼比的识别精度,具有较好的工程应用前景。     

Sentinel-2A数据支持下的雷州半岛植被类型识别

本文以雷州半岛为研究区,利用Sentinel-2A影像数据和真实植被样本数据,综合探讨了机器学习中随机森林与支持向量机的分类效果,并与传统的最大似然法进行比较。提取Sentinel-2A影像9个波段、7个植被指数、72个纹理特征,通过递归特征消除法挑选了10个特征组合,并将其应用于3种分类方法中,对其分类效果进行比较。结果表明:①有效使用多种特征变量是提高植被类型识别精度的关键,就不同特征对植被类型识别的重要性而言,光谱特征与纹理特征相当且大于植被指数,三者重要性相差不大;②随机森林分类效果最佳,不但能对特征进行有效选择,而且能保证植被类型提取精度,提高运行效率;③基于随机森林特征选择的递归特征消除法得到的特征组合不能对其他分类器性能进行优化,对随机森林模型本身的优化效果也有限。     

赤潮灾害处理技术概述

文中介绍了包括卫星遥感、航空遥感、船载和浮标现场观测等监测技术在内的赤潮灾害监测技术;阐述了化学方法、物理方法和生物方法等赤潮灾害消除技术;论述了赤潮灾害损失评估技术;并探讨了赤潮灾害处理技术发展趋势.     

延庆五里营动水位异常变化分析

延庆地震台五里营井动水位2010年1月13日15时48分突然大幅下降,7min下降0．47m,针对此异常变化,台站人员通过检查仪器运行、观测环境等多方面查找原因,逐项排除干扰因素,最后通过科学、系统的分析,判定此异常是由于地下泄流管破裂,造成水位迅速下降,非地震前兆异常,避免了误报地震,为水位异常判断积累了一定经验。     

Is There Any Correlation Between Landscape Characteristics and Total Nitrogen in Wetlands Receiving Agricultural Drainages?

In the case of increasing fragmentation of wetlands, the study of the relationship between wetland landscape characteristics and total nitrogen(TN) in water is of great significance to reveal the mechanism of wetland water purification. Taking the Naoli River(NR) wetlands in Northeast China as the research object, 10 uniformly distributed sampling sites in the study area were sampled in August 2015 to test the TN concentration and interpret the images of NR wetlands in the same period. Taking the sampling site as the control point, the whole wetlands were divided into 10 regions, and the landscape index of each region was extracted. In order to reveal whether the landscape characteristics are related to the TN concentration in the wetlands water body, the landscape index and the TN concentration in the control point water body were analyzed by correlation analysis, step-by-step elimination analysis and path analysis to reveal whether the landscape characteristics are related to the TN concentration under wetlands receiving agricultural drainages. The results showed that the correlation coefficients between four area indexes or eight shape indexes and TN concentration did not reach a significant correlation level(P > 0.05), indicating that TN removal was not only determined by a single landscape index. The path coefficient of edge density(ED) index is –0.41, indicating that wetland patch connectivity is the primary factor of TN removal, and there is no relationship between the larger patch area and the higher TN removal. The removal of TN in wetlands is restricted by the synergistic effect of landscape area and shape characteristics.     

局域网络故障检测技术方法

局域网络故障可分为常规故障和非常规故障。对常规故障应采用检查操作对象、分段检测、逐一排除、最终定位故障源的技术方法;非常规故障是一种既复杂又无规则的常见问题,对此种故障应采取屏幕提示、理论分析、经验判断等相结合的技术手段最终定位故障源。     

波前法在高层结构计算中的应用

概述了高层结构的分析方法，给出了高层结构计算的两种波前法计算过程。文末还给出了广东国际大厦（６３层）模型结构的计算结果，其实测值非常吻合。     

Elimination of Interferences in the Determination of Palladium,Platinum and Rhodium Mass Fractions in Moss Samples using ICP‐MS/MS

Moss samples (Pleurozium schreberi) exposed to traffic‐related emission of Pd, Pt and Rh were analysed in this study. Successful elimination of interferences was achieved in the determination of Pd, Pt and Rh mass fractions in these samples using inductively coupled plasma‐tandem mass spectrometry (ICP‐MS/MS). Based on the results, a reliable determination of Pd, Pt and Rh mass fractions in microwave‐digested moss samples was obtained using ammonia (10% NH₃ in He) in the collision/reaction cell when ¹⁰³Rh was measured either on‐mass (103→103) or with mass‐shift (103→171) and mass‐shifts for ¹⁰⁸Pd (108→159) and ¹⁹⁵Pt (195→229) were used. The ICP‐MS/MS procedure was validated using BCR‐723 (road dust) as a reference material. In addition, a good agreement between the ICP‐MS/MS results and the results obtained with cloud point extraction and quadrupole ICP‐MS was observed for the moss samples.     

二维中值滤波在电磁测深曲线去噪中的应用

作者针对目前点距较密的电磁测深（如CEMP）数据处理问题,提出并应用二维中值滤波方法进行数据处理,通过与传统去噪方法相比较,该方法对于压制或消除实测数据中的随机干扰以及强工业电磁干扰的效果更好,同时对于压制静态位移也有明显的作用。     

地震道的奇性检测与提高分辨率

地震道蕴涵有丰富的奇性特征,而地震子波的起跳点是形成地震道奇性特征的主要原因.样条函数是间段性和连续性的对立统一体.对于地震道,子波起跳点(波至)具有某种意义上的间断性.针对地震道的奇性特征,应用了奇性检测算法,具体过程为:首先以样条函数为尺度函数,构造一个低通滤波器,采用小波包的思想,定义一个算子,使低频分量逐步迭代地从原始信号中分离出,便可以对地震道信号进行奇性分解.经过理论模型和实际资料的处理,能显著提高分辨率,对于面波的去除也很理想;此算法没有任何假设条件的约束.