口服免疫药物后中国对虾某些血淋巴因子的测定及方法研究

于１９９１年４－９月，以海捕中国对虾亲虾及养殖中国对虾为材料，运用单向免疫扩散（ＳＲＩＤ）方法测定对虾血淋巴免疫因子，并采用经改进后的Ｈｏｒｏｗｉｔｚ等人的方法测定对虾体内的酚氧化酶（ＰＯ）活性，目的在于运用该两种方法对口服不同免疫药物的养殖对虾的血淋巴免疫因子及ＰＯ活力进行测定。对亲虾和养殖对虾口服不同免疫药物前后的测定结果表明，几种情况下的对虾血淋巴中均存在类似ＩｇＭ的因子，血细胞中存在酚氧化     

Spatio-Temporal Distribution of Algal and Animal Communities in a Posidonia oceanica Meadow

Abstract. Investigations were carried out in a Posidonia oceanica meadow at Ischia (Gulf of Naples) along a depth gradient from 1 to 32 m in November 1979 and May 1981.
In these two months, different discontinuities were found at the sampling stations (1, 5, 10, 20, 30 m) at 10 m for leaf features, between 1 and 5 m for the algal community and the vagile fauna. The leaf features (length, surface, and biomass) along the transects show a delay in leaf production toward the deeper stations, below the thermocline.
The algal community shows in both seasons and in all stations a persistence of an encrusting layer, mostly represented by Corallinaceae and the brown alga Myrionema orbiculare , while at the shallowest station (1 m) the community is characterized by a more developed upright layer.
The vagile fauna of the leaf stratum, mostly represented by Polychaeta, Mollusca , and Crustacea (Peracarida and Decapoda) , shows a consistent zonation in both seasons along the transect. A superficial community at I m, characterized by a low number of species and individuals and specialized for high environmental energy levels, and a deeper community, persistent in time and more strictly related to the Posidonia oceanica meadows, are identified.
The influence of environmental factors and the importance of meadow structure for the zonation of the algal and animal communities are discussed.     

The Working Mechanism and An Analytic Method of the Deep Embedded Large Cylinder Structure

The large cylinder is a new-type structure that has been applied to harbor and offshore engineering. An analytic method of the relationship between loads and the structure displacement is developed based on the failure mode of deep embedded large cylinder structures. It can be used to calculate directly the soil resistance and the ultirnate bearing capacity of the structure under usage. A new criterion of the large cylinder structure, which discriminates the deep embedded cylinder from the shallow embedded cylinder, is defined. Model tests prove that the proposed method is feasible for the analysis of deep embedded large cylinder structures.     

基于神经网络的平底结构砰击压力预报

对利用神经网络预报平底结构入水砰击压力的方法进行了探讨。首先利用仿真软件计算各种情况下平底结构入水所产生的砰击压力，以此形成训练神经网络的数据集。其次利用数据集对三层反馈式网络进行了训练，讨论了不同隐含层节点数对该非线性系统的拟合能力，并且对梯度下降法、动量修正法和基于优化的LM算法的有效性和精度进行了比较，最后得出了适合平底结构入水砰击系统的网络结构。