摘要：This paper describes one approach to the design of reinforced concrete (RC) bridge piers, using a three-hybrid multi- objective simulated annealing (SA) algorithm with a neighborhood move based on the mutation operator from the genetic algorithms (GAs), namely MOSAMO1, MOSAMO2 and MOSAMO3. The procedure is applied to three objective functions: the economic cost, the reinforcing steel congestion and the embedded CO 2 emissions. Additional results for a random walk and a descent local search multi-objective algorithm are presented. The evaluation of solutions follows the Spanish Code for structural concrete. The methodology was applied to a typical bridge pier of 23.97 m in height. This example involved 110 design variables. Results indicate that algorithm MOSAMO2 outperforms other algorithms regarding the definition of Pareto fronts. Further, the proposed procedure will help structural engineers to enhance their bridge pier designs.

摘要：This paper deals with a design approach of a gait training machine based on a quantitative gait analysis. The proposed training machine is composed of a body weight support device and a cable-driven parallel robot. This paper is focused on the cable-driven robot, which controls the pose of the lower limb through an orthosis placed on the patient＇s leg. The cable robot reproduces a normal gait movement through the motion of the orthosis. A motion capture system is used to perform the quantitative analysis of a normal gait, which will be used as an input to the inverse dynamic model of the cable robot. By means of an optimization algorithm, the optimal design parameters, which minimize the tensions in the cables, are determined. Two constraints are considered, i.e., a non-negative tension in the cables at all times, and a free cable/end-effector collision. Once the optimal solution is computed, a power analysis is carried out in order to size the robot actuators. The proposed approach can be easily extended for the design study of a similar type of cable robots.

摘要：A "file-to-factory" process of computer technology is a way to both maximise efficiency throughout the building process, in crease a buildi ng's performance, and be able to add interesting architectural possibilities throughout the design phase. The authors investigate a novel approach that produces a set of building trajectories rather than a set of buildings, yet yields a series of build-able examples of those trajectories. This paper evaluates how this series of stacked multi-storey timber buildings can be both incorporated within a file-to-factory process, and give rise to creating new innovative solutions throughout the entire design and manufacturing process. This process is applied to a real Swedish project called Zembla. It redefines the notion of sprawl, turning it into a progressive tactics for linking the dty fabric to rural areas. It is a postsustainable file-to-factory-produced timber ground-scraper;soaring above ground and water, suggesting a new way of making city-sized buildings for the future. A plug-in grid-shell structure is designed to contain a minimal amount of timber elements, beams make up the lattice, cross-laminated panels add structural support, surfaces come together to form the living capsules. Having the structure undulate across the topography and touching the ground in as few places as possible uses the dichotomy between landscape and urbanism, bringing the city to the people living in less densified areas. Each living unit is customised to its topological conditions within the grid.

摘要：We consider the problem of displaying commer- cial advertisements on web pages, in the ＂cost per click＂ model. The advertisement server has to learn the appeal of each type of visitor for the different advertisements in order to maximize the profit. Advertisements have constraints such as a certain number of clicks to draw, as well as a lifetime. This problem is thus inherently dynamic, and intimately com- bines combinatorial and statistical issues. To set the stage, it is also noteworthy that we deal with very rare events of in- terest, since the base probability of one click is in the or- der of 10-4. Different approaches may be thought of, rang- ing from computationally demanding ones （use of Markov decision processes, or stochastic programming） to very fast ones. We introduce NOSEED, an adaptive policy learning al- gorithm based on a combination of linear programming and multi-arm bandits. We also propose a way to evaluate the extent to which we have to handle the constraints （which is directly related to the computation cost）. We investigate the performance of our system through simulations on a realistic model designed with an important commercial web actor.

摘要：Au has been loaded （1% wt.） on different commercial oxide supports （CuO, La2O3, Y2O3, NiO） by three different methods： double impregnation （DIM）, liquid-phase reductive deposition （LPRD）, and ultrasonication （US）. Samples were characterised by N2 adsorption at -196℃, high-resolution transmission electron microscopy, selected area electron diffraction, energy dispersive X-ray spectrometry, high-angle annular dark-field imaging （Z-contrast）, X-ray diffraction, and temperature programmed reduction. CO oxidation was used as a test reaction to compare the catalytic activities. The best results were obtained with Au loaded by DIM on the NiO support, with an activity of 7.2 × 10^（-4） molco·gAu^（-1）·s^（-1） at room temperature. This is most likely related to the Au nanoparticle size being the smallest in this catalyst （average 4.8 nm）, since it is well known that gold particle size determines the catalytic activity. Other samples, having larger Au particle sizes （in the 2-12 nm range, with average sizes ranging from 4.8 to 6.8 nm）, showed lower activities. Nevertheless, all samples prepared by DIM had activities （from 1.1 × 10^（-4） to 7.2 × 10^（-4） molco·gAu^（-1）·S^（-1）, at room temperature） above those reported in the literature for gold on similar oxide supports. Therefore, this method gives better results than the most usual methods of deposition-precipitation or co-precipitation.

摘要：目的是建立5种被动的动力减振器对长宽比较大的高层建筑侧移-扭转脉动风振反应控制优化设计的统一方法。文中,在建立了5种被动动力减振器对高层建筑侧移-扭转脉动风振反应控制统一的方程的基础上,导出了被动动力减振器对高层建筑侧移-扭转脉动风振反应控制效果最优等效阻尼比和被动动力减振器最优设计参数的基本公式。通过工程实例的计算与分析,本文比较了5种被动动力减振器对高层建筑侧移-扭转脉动风振反应的控制效果,讨论了动力减振器在楼层上设置位置对动力减振器最优设计参数和控制效果的影响,得到了2个有工程价值的重要结论。