基于深度强化学习的TSN流排序和调度

针对现有研究寻找数据流最佳排序这一过程本身花费时间过多,没有工程可行的寻找最佳排序的有效方法的问题,提出基于深度强化学习的流排序和调度框架PSNDRL.该框架包括3个关键模块,即创建时间触发(Time-Triggered,TT)流之间关系图的预处理模块、挖掘和量化TT流之间复杂的相关关系并选择概率值最高的TT流的代理模块、进行TT流调度和奖励计算的环境模块,利用图卷积网络和强化学习从大量的TT流中智能探索流特征以及流之间的复杂相关关系对基于可满足性模理论(Satisfiability Modulo Theories,SMT)流调度算法求解时间的潜在影响.通过训练该框架,学习得到一个高效率的TT流策略排序网络,用于在利用SMT流调度算法调度TT流时进行TT流选择.通过与随机排序和基线排序方法进行对比,验证PSNDRL的有效性.结果表明:与随机排序方法的总调度时间、基线排序方法的最大总调度时间相比,PSNDRL的总调度时间分别减少了24.63%和25.95%.所提框架为提高时间敏感网络(Time Sensitive Networking,TSN)流调度效率的研究提供了新的方向.

基于内容感知的可分级视频码流排序方法

为了提高传输效率,改善视频播出质量,提出一种基于内容感知的可分级视频码流排序方法。根据人眼对运动变化场景的主观感受,自适应加权前、背景运动矢量,建立描述视频动态特性的模型,结合视频编码结构和率失真模型,对帧级别的数据包重要性进行标记并按照优先级进行排序。实验结果表明,该方法能准确描述视频动态性,有效提高视频播出质量,尤其对于运动场景复杂的视频,该方法能明显改善人眼主观感受。

列流图中列流线生成与排序优化算法

针对列流线铺画中枢纽车站数量多、车站衔接方向多以及列车种类多等现实难题,提出列流线生成与排序优化算法,实现列流图的快速铺画。根据列流图“横平竖直”的表现形式,按线路经纬向分割同一方向衔接多条线路的车站,作为后续算法的基础;基于双向扫描算法思想,构建列流线路径的优化生成算法,在任意车站对之间生成列流线备选集合,通过人机交互方式快速确定列流线径路;以列流线交叉点最少兼顾中心对称为优化目标,提出列流线排序优化算法,快速确定列流线在沿途各区间及车站内的空间相对位置。基于列流线生成与排序优化算法开发列流图优化编制辅助设计系统,并以金华铁路枢纽为例进行验证。结果表明:生成列流图无列流径路以及数据标识错误,且列流线备选集合生成以及列流线排序的时间消耗均小于1 s,列流线生成与排序优化算法具有合理性与有效性。

不可分流网络的最小费用流问题

通常情况下单位流量费用最小的那条路径发送各个流总费用是最小的,但是往往单位流量费用最小的那条路径并不一定能满足所有流均可通过。针对不可分流的网络流最小费用问题,提出按流值排序寻求最优解的算法,并给出相关的理论证明及算法,最后通过具体实验测试了该算法的有效性。此算法可以快速求解所提的问题,并能够算出最优值。实例结果表明,该算法有效地解决了不可分流的网络流最小费用问题,可以应用于实际的网络优化中。

面向查询的自动文本摘要技术研究综述

对面向查询的自动文本摘要技术进行系统梳理,分析所用方法的基本思想、优缺点,并总结未来的发展方向。通过分析梳理,总结出了四大类面向查询的自动文本摘要技术:基于图模型的方法、基于机器学习的方法、基于聚类的方法和其他方法。在今后的研究过程中,基于神经网络和多模型融合的方法将成为未来研究的热点,在应用层面上,与实际应用场景相结合的算法研究将成为趋势。

区域进场扇区点融合飞行程序应用探讨

根据点融合飞行程序设计的概念及方法研究点融合飞行程序在区域进场扇区的应用前景,借助AirTOP仿真平台通过对设计方案进行仿真及模拟评估,验证了点融合飞行程序不仅可应用于区域进场扇区,还能有效提升进港航班流的效率。

HEURISTIC STUDY OF FLOWSHOP SCHEDULING TO MINIMIZE MEAN FLOW TIME WITH LOT TRANSFER CONSIDERED

Lot scheduling problem with idle time transfer between processes to minimize mean flow time is very important because to minimize mean flow time is to minimize work in process. But the problem is NP hard and no polynomial algorithm exists to guarantee optimal solution. Based the analysis the mathematical structure of the problem, the paper presents a new heuristic algorithm. Computer simulation shows that the proposed heuristic algorithm performs well in terms of both quality of solution and execution speed.

An integer multi-objective optimization model and an enhanced non-dominated sorting genetic algorithm for contraflow scheduling problem

To determine the onset and duration of contraflow evacuation, a multi-objective optimization(MOO) model is proposed to explicitly consider both the total system evacuation time and the operation cost. A solution algorithm that enhances the popular evolutionary algorithm NSGA-II is proposed to solve the model. The algorithm incorporates preliminary results as prior information and includes a meta-model as an alternative to evaluation by simulation. Numerical analysis of a case study suggests that the proposed formulation and solution algorithm are valid, and the enhanced NSGA-II outperforms the original algorithm in both convergence to the true Pareto-optimal set and solution diversity.

Concentration–discharge patterns of weathering products from global rivers

Quantifying the functional relationships relating river discharge and weathering products places key constraints on the negative feedback between the silicate weathering and climate. In this study we analyze the concentration–discharge relationships of weathering products from global rivers using previously compiled time-series datasets for concentrations and discharge from global rivers. To analyze the nature of the covariation between specific discharge and concentrations, we use both a power law equation and a recently developed solute production equation. The solute production equation allows us to quantify weathering efficiency, or the resistance to dilution at high runoff, via the Damkohler coefficient. These results are also compared to those derived using average concentration–discharge pairs.Both the power law exponent and the Damkohler coefficient increase and asymptote as catchments exhibit increasingly chemostatic behavior, resulting in an inverse relationship between the two parameters. We also show that using thedistribution of average concentration–discharge pairs from global rivers, rather than fitting concentration–discharge relationships for each individual river, underestimates global median weathering efficiency by up to a factor of ～10. This study demonstrates the utility of long time-series sampling of global rivers to elucidate controlling processes needed to quantify patterns in global silicate weathering rates.

Tri-level programming model for combined urban traffic signal control and traffic flow guidance

In order to balance the temporal-spatial distribution of urban traffic flow, a model is established for combined urban traffic signal control and traffic flow guidance. With consideration of the wide use of fixed signal control at intersections, traffic assignment under traffic flow guidance, and dynamic characteristics of urban traffic management, a tri-level programming model is presented. To reflect the impact of intersection delay on traffic assignment, the lower level model is set as a modified user equilibrium model. The middle level model, which contains several definitional constraints for different phase modes, is built for the traffic signal control optimization. To solve the problem of tide lane management, the upper level model is built up based on nonlinear 0-1 integer programming. A heuristic iterative optimization algorithm(HIOA) is set up to solve the tri-level programming model. The lower level model is solved by method of successive averages(MSA), the middle level model is solved by non-dominated sorting genetic algorithm II(NSGA II), and the upper level model is solved by genetic algorithm(GA). A case study is raised to show the efficiency and applicability of the proposed modelling and computing method.

Length-Sorted Semiconducting Carbon Nanotubes for High-Mobility Thin Film Transistors

We have developed a process for chemical purification of carbon nanotubes for solution-processable thin-film transistors （TFTs） having high mobility. Films of the purified carbon nanotubes fabricated by simple drop coating showed carrier mobilities as high as 164 cm2V-1s -1 normalized transconductances of 0.78 Sm-1 and on/off current ratios of 10^6. Such high performance requires the preparation of a suspension of micrometer-long and highly purified semiconducting single-walled carbon nanotubes （SWCNTs）. Our purification process includes length and electronic-type selective trapping of SWCNTs using recycling gel filtration with a mixture of surfactants. The results provide an important milestone toward printed high-speed and large-area electronics with roll-to-roll and ink-jet device fabrication.

Magnetic cell sorting and flow cytometry sorting methods for the isolation and function analysis of mouse CD^4+ CD25^+ Treg cells

Objective: In this paper we compared the two methods of cell sorting (magnetic cell sorting and flow cytometry sorting) for the isolation and function analysis of mouse CD4+ CD25+ regulatory T (Treg) cells, in order to inform further studies in Treg cell function. Methods: We separately used magnetic cell sorting and flow cytometry sorting to identify CD4+ CD25+ Treg cells. After magnetic cell separation, we further used flow cytometry to analyze the purity of CD4+ CD25+ Treg cells, trypan blue staining to detect cell viability, and propidium iodide (PI) staining to assess the cell viability. We detected the immune inhibition of CD4+ CD25+ Treg cells in the in vitro proliferation experiments. Results: The results showed that compared to flow cytometry sorting, magnetic cell sorting took more time and effort, but fewer live cells were obtained than with flow cytometry sorting. The CD4+ CD25+ Treg cells, however, obtained with both methods have similar immunosuppressive capacities. Conclusion: The result suggests that both methods can be used in isolating CD4+ CD25+ Treg cells, and one can select the best method according to specific needs and availability of the methodologies.

Multi-objective optimization of the streamlined head of high-speed trains based on the Kriging model

As the running speed of high-speed trains increases, aerodynamic drag becomes the key factor which limits the further increase of the running speed and energy consumption. Aerodynamic lift of the trailing car also becomes the key force which affects the amenity and safety of the train. In the present paper, a simplified CRH380A high-speed train with three carriages is chosen as the model in order to optimize aerodynamic drag of the total train and aerodynamic lift of the trailing car. A constrained mul- ti-objective optimization design of the aerodynamic head shape of high-speed trains based on adaptive non-dominated sorting genetic algorithm is also developed combining local function three-dimensional parametric approach and central Latin hypercube sampling method with maximin criteria based on the iterative local search algorithm. The results show that local function parametric approach can be well applied to optimal design of complex three-dimensional aerodynamic shape, and the adaptive non-dominated sorting genetic algorithm can be more accurate and efficient to find the Pareto front. After optimization the aerodynamic drag of the simplified train with three carriages is reduced by 3.2%, and the lift coefficient of the trailing car by 8.24%, the volume of the streamlined head by 2.16%; the aerodynamic drag of the real prototype CRH380A is reduced by 2.26%, lift coefficient of the trailing car by 19.67%. The variation of aerodynamic performance between the simplified train and the true train is mainly concentrated in the deformation region of the nose cone and tail cone. The optimization approach proposed in the present paper is simple yet efficient, and sheds lights on the constrained multi-objective engineering optimization design of aerodynamic shape of high-speed trains.

Multi-objective optimal configuration of current limiting strategies

With the level of short-circuit current of power systems growing increasingly higher,optimal allocation of current limiters has received considerable attention in recent years,especially in China.This paper analyzes two kinds of common used current limiters based on the increment of bus impedance matrix and proposes a multi-objective current limiters configuration model considering the investment of limiters,the level of short-circuit current,as well as the transient stability of power system.An innovative search space reduction technique based on sensitivity factor is introduced to choose better candidate locations for current limiters so as to avoid the curse of dimensionality.The elitist non-dominated sorting genetic algorithm II is used to search the Pareto-optimal solutions of the proposed model.In order to further improve optimization efficiency,master-slave parallel modification of NSGA-II program structure is implemented.The satisfactory case study results demonstrate the feasibility of the proposed multi-objective method in power system current limiters allocation associated with cost,security and stability.

Multi-objective pigeon-inspired optimization for brushless direct current motor parameter design

Pigeon-inspired optimization(PIO) is a new swarm intelligence optimization algorithm, which is inspired by the behavior of homing pigeons. A variant of pigeon-inspired optimization named multi-objective pigeon-inspired optimization(MPIO) is proposed in this paper. It is also adopted to solve the multi-objective optimization problems in designing the parameters of brushless direct current motors, which has two objective variables, five design variables, and five constraint variables. Furthermore, comparative experimental results with the modified non-dominated sorting genetic algorithm are given to show the feasibility, validity and superiority of our proposed MIPO algorithm.