基于遗传算法的低轨天基雷达星座设计

从目标监视性能的角度出发,分析了低轨天基雷达星座的卫星总数、能量需求以及星座发射费用等优化目标。综合考虑了Walker星座的碰撞等约束问题,建立了低轨天基雷达星座设计的优化目标。提出了一种基于遗传算法的低轨天基雷达星座设计方法。根据天基雷达星座的特点,该算法在进化过程中使用了非可行解修正和约束分级等约束处理方法,有效提高了算法的运算效率。通过仿真实验验证了该方法能够有效的同时优化低轨天基雷达星座的构形和卫星参数。     

一种新型的低轨存储转发通信星座设计方法

针对低轨存储转发星座的特点以及传统星座设计方法无法同时优化覆盖性能和网络性能的不足,该文提出一种新型的兼顾覆盖性能以及网络性能的低轨存储转发星座设计方法,并采用非劣分层多目标进化算法(NSGA-II)优化该多目标多约束星座设计问题。通过基于STK和OPNET的协同优化仿真平台验证了该星座设计方法的有效性。     

通信卫星星座优化设计综述

通信卫星星座优化设计是构建通信星座系统的关键环节。对通信卫星星座设计中的优化模型和优化算法进行了综述和展望。首先,建立了通信卫星星座设计的一般优化模型;然后,从覆盖性能、星间链路和系统成本三个方面对优化模型中优化变量、目标函数和约束条件的构建进行了综述;之后,介绍了卫星星座设计中常用的三大类多目标优化算法,归纳总结了4种主流算法的主要特征,并比较分析了它们应用于卫星星座设计的优缺点;最后,指出了现有通信卫星星座优化设计中存在的问题及发展趋势。     

采用遗传算法的低轨区域通信星座优化设计

给出一种利用改进遗传算法实现低轨区域通信星座优化设计的方法,有效克服了上述困难。首先建立通用的区域覆盖星座模型,确定优化控制参数,并结合低轨星座应用背景对其进行约束。然后给出一种基于网格点统计的星座性能评价准则。为了提高遗传算法对局部最优解的搜索能力,文章提出一种混合遗传算法,该方法在基本遗传算法中加入复形调优算法,并根据优秀个体的分布对参数区间进行调整。将该算法应用于星座模型,建立一套完整的星座优化设计方案。最后对具体实例进行优化仿真,结果表明该方法取得良好的优化结果。     

一种区域性覆盖星座设计与仿真方法

主要研究了对地面目标进行连续覆盖的星座设计问题。首先分析了星座最小仰角与星下点轨迹、轨道高度的关系,然后推导了对覆盖性能有重要影响的星间覆盖间隔时间和轨道面覆盖间隔时间的计算公式,提出理论算法与STK仿真相结合优化星座设计的方法,给出适合于中国的非静止轨道星座方案。仿真结果表明,该星座能对中国实施24h不间断覆盖通信。     

区域覆盖共地面轨迹星座的优化设计

该文讨论了使用中轨共地面轨迹星座建立区域覆盖卫星通信系统的优化设计方法。在深入研究共地面轨迹星座参数特性的基础上,提出了一种能够完整描述该类型星座的编码标识方法,推导了星座与星座的等价关系,讨论了利用该类型星座实现区域覆盖时的优化方法,给出了适用于我国的多个优化星座方案实例。从仿真结果可以看出,该类型星座能够以较少数量的卫星为我国提供良好的覆盖性能,适于我国的区域覆盖卫星通信系统采用。     

面向低轨巨型星座的多目标卫星接入优化

低轨巨型星座可以提供全球无缝覆盖的卫星通信服务,但也会导致用户被卫星多重覆盖,如何选择最优的卫星进行接入,成为低轨巨型星座接入技术的关键问题。对此,提出一种基于粒子群的多目标卫星接入优化算法。综合考虑星地距离、卫星剩余可服务时间和卫星剩余负载等目标函数,根据不同业务类型的QoS需求,对参数赋予相应的权值;针对可视卫星数量多且动态变化的特点,通过粒子群算法进行低复杂度求解。仿真表明,相较于对比算法,所提算法对星地距离和卫星剩余可服务时间进行了优化折中,保障了服务时间和通信质量,提高了接入成功率,能够灵活满足用户的不同业务类型需求,适应用户业务不同时空分布以及业务离散化的场景。     

区域覆盖卫星系统的共地面轨迹星座实现

研究了使用中轨共地面轨迹星座实现区域覆盖卫星通信系统的优化方法,提出了一种能够完整描述该类型星座的编码标识方法,给出了多个适用于我国的星座方案实例。基于时变的业务分布密度特性对星座方案的参数进行了调整,使得星座的覆盖特性与业务分布密度更加一致。仿真结果表明该类型星座能够以较少数量的卫星为我国提供良好的覆盖性能。     

一种基于MOSA的低轨卫星星座多波束分配策略

针对LEO宽带卫星星座中多卫星同时覆盖某一区域情况下的多波束资源分配问题,采用地面固定波束体制,研究了多卫星多波束的分配策略。以地面用户接入容量需求和端到端时延最小化为优化目标,通过对多目标模拟退火算法进行改进优化,建立了一种可同时满足用户容量与时延要求的多卫星多波束的分配方法。仿真结果表明,算法能够根据用户优先级,有效提升星座波束的通信与覆盖性能,通信时延降低了4.1%～20.6%。     

基于工艺参数扰动的IC参数成品率多目标优化算法

在芯片制造工艺中,参数扰动影响了集成电路（Integrated Circuit,IC）成品率,使不同参数成品率间存在着此消彼长的相互制约关系,而目前IC参数成品率优化算法却主要局限于单一优化目标问题。本文提出一种基于工艺参数扰动的参数成品率多目标优化算法。该算法针对漏电功耗成品率及芯片时延成品率,首先构建具有随机相关性的漏电功耗及芯片时延统计模型;随后根据其相互制约特性建立基于切比雪夫仿射理论的参数成品率多目标优化模型;最后利用自适应加权求和得到分布均匀的帕雷托优化解。实验结果表明,该算法对于具有不同测试单元的实验电路均可求得大约30个分布均匀的帕雷托优化解,不仅能够有效权衡多个优化目标间的相互制约关系,还可以使传统加权求和优化方法在帕雷托曲线变化率较小之处得到优化解。     

基于蚁群算法的雷达海杂波反演蒸发波导研究

蚁群算法是近年来兴起的一种随机搜索寻优算法,为函数优化问题提供了一种新思路;利用雷达海杂波反演蒸发波导(RFC)可以提供近实时的波导信息,为雷达等无线电系统评估提供参数,且无需增加硬件设备.阐述了利用雷达海杂波反演大气波导的基本原理和基本流程;给出了用于函数优化的蚁群算法基本原理;将蚁群算法与雷达海杂波反演大气波导技术相结合实现了对蒸发波导的反演.利用我国东部沿海的实验数据对反演算法进行检验,分析结果显示蚁群算法收敛速度较快,精度较高.     

基于组件式蚁群算法的车辆路径问题研究

现有方法仅从蚁群算法的基本结构出发设计软件,缺少软件设计模型的有效指导,很难用来求解不同种类的优化问题.鉴于此,应用组件软件设计方法,提出了一种组件武蚁群算法.该方法力求在结构上直接反映蚁群的本质思想和关键概念;最大程度降低与问题的相关性;强调以接口为中心的设计理念.大量车辆路径问题的验证结果表明,组件式蚁群算法性能优良,能够有效地求解车辆路径问题.该方法易于理解和使用,具有很强的可重用性和可扩展性,为求解各类优化问题提供了很好的起点和可持续发展的框架.     

Theoretical analysis of RF performance of anisotropic conductive adhesive flip- chip joints

The RF performance of a single flip-chip joint using anisotropic conductive adhesive (ACA) was modeled using three-dimensional finite difference in time domain (FDTD) methodology. The effects of the number, the distribution of ACA conductive particles, as well as the ACA resin, on RF transition of the joint were studied. The results show that the number and the distribution of conductive particles as well as the adhesive resin have limited influence on the RF performance of an ACA flip-chip joint. Based on the FDTD simulation results, an equivalent circuit model of a single ACA flip-chip joint was also proposed.     

基于连续空间优化的蚁群算法

通过对蚁群算法基本理论的研究,从经典的蚁群算法模型中,抽象出解决问题的一般方法,提出了在连续空间优化问题中蚁群算法的模型,在算法中加入了自适应策略用以提高算法的性能,并通过实例分析了连续空间优化问题中蚁群算法的性能,通过仿真实验证明了算法的可行性与实用性。     

Experimental and theoretical characterization of electrical contactin anisotropically conductive adhesive

Electrical conduction through anisotropically conductive adhesive (ACA) is caused by deformation of metal fillers under pressure and heat. In this work, the hardness of the electrical particles under various deformation degrees was determined by nano-indentor measurements and the electrical resistance of the electrical contacts was measured under various deformation degrees. Theoretical model and simulation have been developed for the microscopic mechanism of the electrical conduction through metal fillers in the anisotropically conductive adhesive. By comparing with experimental data it is concluded that the deformation of the metal filler in our ACA is plastic even at rather low external load. Further theoretical simulation reveals two important aspects of the conductance characteristics. The conductance is improved by increasing the external load but the dependence of the conductance on the spatial position of the metal filler becomes stronger. Design and optimization of the ACA with respect to the absolute value of the electric conductance and its dependence on the spatial position of the metal filler are of essential importance for the electronics packaging application of the anisotropically conductive adhesives     

Using Data Mining to Find Patterns in Ant Colony Algorithm Solutions to the Travelling Salesman Problem

Travelling Salesman Problem （TSP） is a classical optimization problem and it is one of a class of NP- Problem. The purposes of this work is to apply data mining methodologies to explore the patterns in data generated by an Ant Colony Algorithm （ACA） performing a searching operation and to develop a rule set searcher which approximates the ACA＇s searcher. An attribute - oriented induction methodology was used to explore the relationship between an operations＇ sequence and its attributes and a set of rules has been developed. At the end of this paper,the experimental results have shown that the proposed approach has good performance with respect to the quality of solution and the speed of computation.     

Investigation of Mechanical and Electrical Characteristics for Cracked Conductive Particle in Anisotropic Conductive Adhesive (ACA) Assembly

In an anisotropic conductive adhesive (ACA) assembly, the electrical conduction is usually achieved with the conductive particles between the bumps of integrated circuit (IC) and corresponding conductive tracks on the glass substrate. Fully understanding of the mechanical and electrical characteristics of ACA particles can help to optimize the assembly process and improve the reliability of ACA interconnection. Most conductive particles used in the ACA assembly are with cracks in the metal coating of the particles after the ACA bonding. This paper introduced the fracture analysis by applying the cohesive elements in the numerical model of the nickel-coated polymer particle and further simulating the cracks initiation and propagation in the nickel coating during the ACA bonding. The simulation results showed that the stress distribution on the nickel-coated particle with cracks was significantly different from that on the nickel-coated particle without crack, indicating that the stress analysis by taking the crack into consideration is very important for the reliability assessment of the ACA interconnection. The stress analysis of cohesive elements indicated that the cracks initiated at the central area of the nickel coating and propagated to the polar area. Furthermore, by the introduction of a new parameter of the virtual resistance, a mathematical model was established to describe the electrical characteristics of the nickel-coated particle with cracks. The particle resistance of the nickel-coated particle with cracks was found to be much higher than that of the particle without crack in the optimized bonding pressure range, indicating that it is necessary to take the crack into consideration for the particle conduction analysis as well. Therefore, the fracture analysis on the conductive particle by taking the crack into consideration could accurately evaluate the reliability of ACA interconnection and avoid serious reliability issues.     

Formulation and characterization of anisotropic conductive adhesive paste for microelectronics packaging applications

There has been a steadily increasing interest in using electrically conductive adhesives as interconnecting materials in electronics manufacturing. In this paper, several anisotropic conductive adhesive (ACA) pastes were formulated, which consist of diglycidyl ether of bisphenol F or diglycidyl ether of bisphenol A as polymer matrix, imidazoles as curing agents, and different sizes of silver (Ag) powders or gold (Au)-coated polymer spheres as conductive particles. The effects of ACA resin and different curing agents, as well as different conductive particles, on flexible substrate of the flip-chip joint were studied. The results show that the size and type of different conductive particles have very limited influence on an ACA flip-chip joint. The ACA resin as well as the curing agent can affect the reliability of the joint. The same results can be applied for the failure analysis of ACA flip-chip technology.     

一种新的混合并行蚁群算法研究应用

提出了一种新的混合并行蚁群算法,在单机多核机及多核集群机下分别实现了MPI并行蚁群算法及MPI+TBB并行蚁群算法,应用于真实路网车辆路径问题(Vehicle Routing Problem,VRP),对两者进行了实验对比,实验结果表明MPI并行蚁群算法具有较高的加速比,和问题规模关系不大,刚开始呈现线性加速比。较MPI并行蚁群算法,MPI+TBB混合并行蚁群算法具有更好的可扩展性,在进程数较多时仍具有较高的加速比。     

一种基于蚁群算法的多目标跟踪数据关联方法

针对多目标跟踪问题,提出基于蚁群算法的数据关联方法.首先将多目标跟踪问题描述为组合优化问题.利用蚁群算法解决组合优化问题的优势,在对其路径和路径长度概念重新定义的基础上,将其应用于选择多目标跟踪中的轨迹-观测关联集合.详细介绍了蚁群算法应用于数据关联问题的具体方法,建立了基于蚁群算法的数据关联模型.实验结果表明,论文提出的基于蚁群算法的数据关联方法是行之有效的.