基于SVM的海底声纳图像底质识别*

对海底底质声纳图像分割,提取单元特征向量进行主成份分析,选取均值、标准差、对比度、相关系数、能量及同质性作为训练特征向量,优化支持向量机(SVM)参数.利用多分类 SVM 对砂、礁石、泥进行训练,有效、快速地分离出了3种底质类型,测试精度高于80%,表明该方法可有效用于海底底质声纳图像的识别和分类.     

卫星海洋遥感数据的无损压缩及解压算法

在分析卫星海洋遥感单轨三级数据产品特征的基础上,针对游程编码无损压缩算法特征,提出了优化游程编码压缩算法,运用该算法实现了海洋遥感三级数据产品的无损压缩,并与流行的WinRAR压缩软件进行了比较测试,结果表明上述算法对海洋遥感数据均有较高的压缩率,优化游程编码压缩算法在空间复杂度及时间复杂度上均有明显的比较优势,可极大节约海洋遥感数据的存储与共享发布空间。     

�Ľ�����Ⱥ�㷨���GPS�̻�������ģ���ȵ��о�

????????????????????????????????????????????????IPSO??????????????·?????IPSO????????????????????????????б????????????????????????????????????????????????PSO?????????????????????????????????????????????????LAMBDA????????????????????·??????п??????????????????????????????????ü????     

Strain and culture medium optimization for production enhancement of prodiginines from marine-derived Streptomyces sp. GQQ-10

A mutant (GQQ-M6) of a Sponge-Derived streptomyces sp. GQQ-10 obtained by UV-induced mutation was used for producing prodiginines (PGs). Single factor experiments and orthogonal array design (OAD) methods were employed for medium optimization. In the single factor method, the effects of soluble starch, glucose, soybean flour, yeast extract and sodium acetate on PGs production were investigated individually. In the subsequent OAD experiments, the concentrations of these 5 key nutritional components combined with salinity were further adjusted. The mutant strain GQQ-M6 gave a 2.2-fold higher PGs production than that of the parent strain; OAD experiments offered a PGs yield of 61mg L-1, which was 10 times higher than that of the initial GQQ-10 strain under the original cultivation mode.     

歧口凹陷深层地震岩石物理特征分析及应用

为明确歧口凹陷深层的岩石物理及AVO响应特征,分析歧口地区全波测井资料,优选该区敏感岩石物理参数;根据分岩性拟合求取纵横波转换经验公式,并采用流体置换原理进行叠前地震道集正演后提取AVO属性.结果表明:泊松比、纵横波速度比为该区有利储层预测的敏感参数,含气砂岩与含水砂岩的数值范围差异明显;歧口深层AVO响应特征可划分为3类:第一类是与气层相对应的反射振幅随着偏移距的增大而负向急剧增大;第二类是与油气同层相对应的反射振幅随着偏移距的增大而负向缓慢增大;第三类是与油水层及干层相对应的反射振幅随着偏移距的增大而负向减小.该结果可为歧口凹陷深层开展流体预测可行性提供依据.     

Design principle for parameters and interferences of tungsten carbide insert interface on roller bits

The lifetime of roll-bits with inserted tooth will be obviously decreased while drilling in strata and the tungsten carbide can be integrally separated from the roll-cone,which will need more trip out time to replace the roller bit.This study provides an entire analysis on stress as well as strength for teeth and tooth holes in loaded conditions.Qualities of tooth drop from matrix within roller bit.The optimized interferences of teeth inserted into the bit can be reasonably obtained by the critical interference derived from some relevant influence factors on the tooth-fixing quality of roller bits.     

基于水环境承载力的沁河流域系统优化调控

基于水环境承载力核算模型,提出流域系统优化调控方法,并应用于沁河流域。通过承载力核算模型计算出2015年和2020年流域水环境承载力,当系统动力学预测的2015年和2020年经济发展规模高于承载力时,通过约束条件中工业企业排放浓度、城镇生活污水收集率、水资源利用率、尾水回用率等4项调控指标进行敏感度分析,最终确定最优的经济发展。结果表明:沁河流域应该优先调控城镇生活污水收集率;2015年,执行工业污水排放二级标准,收集率须达到71.4%以上,优化后的经济规模阈值为475.6亿元;2020年,执行工业一级标准,城镇生活污水收集率须为78.4%以上,相应地经济规模阈值为816.0亿元。     

基于太平洋海温的冬小麦春季湿渍害预测模型

为了进一步开展冬小麦春季湿渍害的预测预报业务工作,利用江苏省1969-2008 年60 个气象站常规观测资料和冬小麦产量资料,根据场相关分析原理,将江苏省春季湿渍害3 个风险区的湿渍害指数与太平洋海温资料进行相关普查,结果表明:3 个风险区的小麦春季湿渍害指数和西太平洋海温均存在高相关区,主要位于Nino 区和西太平洋北部海区,其相关程度随时间具有一定的变化趋势;挑选出与湿渍害指数相关的强信号海区作为预报因子,通过最优化相关处理提高预报因子相关性;经稳定性和独立性检验,创建了对应3 个风险区的冬小麦湿渍害指数预测模型,所有模型均通过0.01 显著性水平检验,拟合及试报结果理想,表明预测模型具有一定应用价值。     

中国石化产业空间组织的评价与优化(英文)

The spatial organization of the Chinese petrochemical industry was optimized ac-cording to the status of development of the industry employing linear programming and Ar-cGIS spatial analysis tools. We first identified the indexes of the spatial organization of the petrochemical industry and established a comprehensive evaluation index system that in-cludes four major categories and 11 indicators. The weight of each index was then deter-mined by the analytical hierarchy process. Afterward, taking the 337 Chinese prefecture-level administrations as basic units and scientifically evaluating the potential comprehensive layout coefficients of the cities, 151 prefecture-level administrative units were selected as the basis for the choice of optimization sites with a linear programming model. Secondly, using the 151 prefecture-level administrative units and the maximum-coverage model, the optimal number and spatial distribution of refineries were identified for service radii of 100, 200 and 300 km. Thirdly, considering the actual distribution of China’s refineries, general rules for the number of refinery layout points and objective values were summarized, and 52 refinery layout points were selected for China. Finally, with ArcGIS spatial analysis tools, the spatial effect of the 52 optimal refinery layout points was simulated for the service scope and socioeconomic factors respectively, and the GDP and population data for each refinery layout point were then ex-tracted within the service scope. On this basis and with estimation of the intensity of crude-oil consumption, final results were obtained for the optimal spatial organization of the Chinese refining capacity and ethylene production capacity.     

塔克拉玛干沙漠沙尘演变及气候因素分析

Based on the sand dust storms data and climatic data in 12 meteorological stations around sand dust storm originating areas of the Taklimakan Desert, we analyzed the trends of the number of dust storm days from 1960 to 2005 as well as their correlations with temperature, precipitation, wind speed and the number of days with mean wind speed 〉 5 m/s. The results show that the frequency of dust storm events in the Taklimakan region decreased with the elapse of time. Except Ruoqiang and Minfeng, in the other 10 meteorological stations, the frequency of dust storm events reduces, and in 4 meteorological stations of Kuqa, Korla, Kalpin and Hotan, the frequency of dust storm events distinctly decreases. The temperature has an increasing trend, while the average wind speed and the number of days with mean wind speed ≥ 5 m/s have decreasing trends. The correlation analysis between the number of days of dust storms and climatic parameters demonstrates that wind speed and the number of days with mean wind speed 〉 5 m/s have strong positive correlation with the number of days of dust storms, with the correlations coefficients being 0.743 and 0.720 （p〈0.01）, respectively, which indicates that strong wind is the direct factor resulting in sand dust storms. Whereas precipitation has significant negative correlation with the number of days of dust storms （p〈0.01）, and the prior annual precipitation has also negative correlation, which indicates that the prior precipitation restrains the occurrence of sand dust storms, but this restraining action is weaker than the same year＇s precipitation. Temperature has negative correlation with the number of dust storm days, with a correlations coefficient of -0.433 （p〈0.01）, which means that temperature change also has impacts on the occurrence of dust storm events in the Taklimakan region.