箱型截面的齐次广义屈服函数与结构极限承载力

针对目前箱型截面广义屈服函数为非齐次函数,采用弹性模量调整法求解箱型结构极限承载力时易出现计算结果受荷载初始值影响和计算精度受损的问题,建立了箱型截面的齐次广义屈服函数,并通过误差分析确定了齐次多项式的阶次,进而定义了箱型截面构件的单元承载比、承载比均匀度和基准承载比,建立了以单元承载比为基本参数的弹性模量调整策略,据此提出了箱形结构极限承载力分析的弹性模量缩减法;并通过算例对该方法进行验证,算例分析表明,该方法计算箱型截面结构极限承载力时,结果不受荷载初始值的影响,具有良好的计算精度和迭代稳定性.     

泛代数上的Groebner-Shirshov基理论

综述了域上或交换代数上的线性（-）代数的相应的簇（范畴）的 Groebner-Shirshov 基理论的新成果，如:结合代数（包括群（半群）代数），自由代数的张量积，李代数，Di-代数，pre-李代数，Rota-Baxter代数，metabelian李代数，L-代数，半环代数，范畴代数，等．以上结果包含了许多应用，尤其是给出了一些著名结论的新的证明．     

基于异步动态的多机器人系统的一致性

研究通过无线网络进行通信的多机器人系统的一致性问题。多机器人系统在固定通信拓扑下可建模为特定的网络控制系统,考虑信道噪声、不确定短时延和数据包丢失等因素的影响时,其可进一步视作由丢包率约束的异步动态系统。通过构造李雅普诺夫(Lyapunov)函数,运用线性矩阵不等式(linear matrix inequality, LMI)方法,得到了该系统指数稳定的充分条件。为实现多机器人系统的运动一致性,设计了控制器；讨论了不同的指数收敛率对多机器人系统收敛速度的影响,并进行了相应的理论证明和仿真验证。     

Social Media as a Mean for Improved Technical Communication

Technical communication (TC) is an important activity in order to provide the users of technical artifacts with necessary information concerning the operation and maintenance of the products they are using. In the current working procedures of TC, however, the users become too passive. They are just receivers of information. Further, there is no effective feedback from the users to the producers of the information. In order to overcome those problems we here propose an application of social media (SM) in future TC. By using SM as a complementary channel for TC it becomes possible for technology firms to boost their competitive advantage and to improve the quality and completeness of their TC. This technique, however, is still more of a potential and less of a reality. Several problems, mostly concerning security and confidentiality remain to be solved. The empirical base for this work comes from TIC, an EU-funded development project involving several Swedish technology companies. Published research results are somewhat scattered but several sources coherently indicate both the need and potential for SM as a vehicle in TC. By scanning the net it is further possible to identify a number of seemingly successful applications of SM in such applications.     

Reproductive ecology of the glass frog Espadarana prosoblepon (Anura: Centrolenidae) in an urban forest of the Central Andes of Colombia

Population and reproductive behaviour aspects of the glass frog Espadarana prosoblepon are well known for populations from Central America, but similar information is lacking for populations from South America. We recorded the reproductive ecology of a population of the glass frog E. prosoblepon in the city of Armenia, Central Andes of Colombia. With weekly surveys with mark-recapture between December 2013 and November 2015, we monitored activity patterns and evaluated if the probability of males mating is associated to their body size or to chorus tenure (i.e. the number of nights a same individual is calling for attracting a mate). In addition, upon observing an amplexus, we recorded the behaviour of the individuals until oviposition and noted characteristics of egg clutch. We recorded 47 males and 32 females, nine amplexus events, eight ovipositions, and 21 egg clutches (35.4 ± 4.79 eggs per clutch, N = 29). Activity of males and females and the number of egg clutches were positively correlated to rainfall. Mating success of males was not associated to their body size, but it was positively associated to longer chorus tenure; clutches from large females had a higher number of eggs than clutches from small females. Oviposition occurred on average 3.76 ± 1.74 hours after we first observed the amplexus, occurring 93.1% of the time in fronds of Selaginella geniculata at an average height of 1.58 ± 0.44 m. After the oviposition, the male left the site, while the female remained between 30 and 45 min, partially covering the eggs. The reproductive behaviour of E. prosoblepon did not vary widely between a population in Costa Rica and our population in time to oviposition, and mating success of males associated to chorus tenure; however, in our study population we recorded a larger clutch size and the preference for S. geniculata fronds as oviposition substrate.     

基于神经网络动态优化的再入轨迹设计

为解决当飞行器出现破损以及飞行器受干扰时的轨迹优化问题,新一代的再入式飞行器需要有实时在线轨迹优化的能力。神经网络动态优化算法(neural dynamic optimization,NDO)的主要特点是能使神经网络逼近最优解。神经网络动态优化算法可以避免传统的间接法在求解轨迹优化问题时协态变量初值猜测问题。给出了神经网络动态优化的原理,详细介绍了优化流程。仿真结果表明神经网络动态优化算法可以很好的避免协态变量初值猜测问题,具有较强的鲁棒性,能满足实时性要求。