Flow Control Strategy for the High Speed Network Based on Control Theory

Design of an effective congestion control scheme is a hot topic in the development of computer network. The flow control scheme can adjust the packet sending rate in source host, thus effectively avoiding the network congestion. This paper proposes a new flow control scheme based on discrete control theory. The simulation results show that this method can adjust the sending rate and queue level in buffer rapidly and effectively. The method is easy to implement and applicable to high speed networks.     

Calculation of angular glint in near field utilizing graphical electromagnetic computing

The angular glint in the near field plays an important role on radar tracking errors. To predict it more efficiently for electrically large targets, a new method based on graphical electromagnetic computing （GRECO） is proposed. With the benefit of the graphic card, the GRECO prediction method is faster and more accurate than other methods. The proposed method at the first time considers the special case that the targets cannot be completely covered by radar beams, which makes the prediction of radar tracking errors more self-contained in practical circumstances. On the other hand, the process of the scattering center extraction is omitted, resulting in possible angular glint prediction in real time. Comparisons between the simulation results and the theoretical ones validate its correctness and value to academic research and engineering applications.     

虚拟维修中的虚拟人移动仿真

虚拟维修是近年来提出的一个新概念,并在维修性分析、维修训练等领域得到广泛的应用。为了解决维修过程仿真中的维修人员仿真问题,提出了一个分阶段的虚拟人移动仿真方法,采用随机运动规划方法生成虚拟人躯干及头部的无碰撞路径,利用运动捕获数据和运动变形来实现对四肢运动的仿真。在运动规划中引入了姿态转移图的概念,实现了虚拟人移动过程中多种姿势之间的自然转化。采用基于运动变形的方法消除了虚拟人上肢和环境中其他物体的碰撞现象,提高了仿真真实性。

Abstract：

Virtual maintenance is a new concept introduced in the last few years,and has been broadly used in maintainability analysis and maintenance training.In order to solve the virtual human simulation in maintenance process simulation,a multi-stage virtual human locomotion simulation method was proposed.The free collision path of virtual human torso and head was planned by a probabilistic motion planner.And the motion of arms and legs was simulated using motion captured data and motion warping.A new concept of pose transition graph was introduced in order to implement natural pose transition in locomotion simulation.In motion optimization,a method based on motion warping was proposed to avoid the possible collision between virtual human arms and other objects in the environment.     

Performance evaluation from stochastic statecharts repre-sentation of flexible reactive systems：a simulation approach

This paper focuses on the performance analysis of flexible reactive systems. The performance analysis consists of two phases： first system modeling, second performance evalua-tion. The paper models the flexible reactive system by the stochas-tic statecharts method, and uses the simulation method to evalu-ate the performance. To make use of the feature of event-triggered state transitions in the statecharts, a new method of simulation is proposed based on the techniques of the discrete-event system simulation. The new method solves the problem of computer imple-mentation of stochastic events, probabilistic transition, concurrent states, paral el actions, and broadcast communication mechanism in the stochastic statecharts. An example of a flexible manufactur-ing system is presented. The simulation result of the example is consistent with the analytical result, which shows the feasibility of the proposed new simulation method.     

虚拟地球场景的可视化建模与仿真

从地球自身模型出发,提出了按季风带划分大气云层实现其转动的新方法。为了提高拟真性,探讨了大气散射的成因,研究了大气散射算法并通过GPU编程实现,最后给出了基于多重纹理混合映射的星空背景模型。实验结果表明,这些方法可以大幅提高整个场景的拟真度,同时具备高实时性。可以被广泛地用于空间作战、探测等仿真任务。

Abstract：

Starting from the earth model,a new method was proposed that the rotation of atmospheric could be divided by monsoon belt.In order to improve the fidelity,the reason caused atmospheric scattering was investigated,and the scattering algorithm was studied and implemented by GPU programming.At the end,the star scene model based on multiple texture blending was given.The result of experiment shows that these methods have not only given a substantial increase in fidelity,but also guaranteed real-time rendering.It can be widely used in simulation of space battlefield and exploration.     

基于KD-DRT的雷达射频信号半实物仿真系统实现

首先介绍了分布实时仿真系统的四层体系结构,从而引出支撑系统KD-DRT;再提出采用这种分层的分布实时仿真技术构建精确制导武器的雷达射频信号半实物仿真系统;对系统模块的层次化分解以及分布系统的时间同步策略进行了分析,最后给出了系统的具体实现结构,验证了KD-DRT针对雷达信号级分布实时仿真的有效性。

Abstract：

4-tier hiberarchy was put forward as the standard architecture of distributed real-time simulation system, and KD-DRT was introduced as a supporting platform. A radar RF signal simulation system for precisely guided missile was constructed based on KD-DRT. The system modules were arranged to different layer according to its time restriction and the time synchronization method of the distributed system was studied. At last, the simulation system was constructed and the efficiency of KD-DRT in simulation of radar signal was proved.     

基于视景通道的透镜成像过程仿真技术研究

针对计算机成像模型在使用过程中存在的不足,提出了对该模型进行完善与改进的方法。首先根据数字变焦原理,推导出放大倍率与视锥角的关系公式,实现了视景通道的变焦功能,该方法计算量小,效果逼真;其次,提出了一种镜头对焦过程的仿真方法,它充分利用粒子系统技术模拟对焦成像由模糊到清晰的过程,不仅获得更真实的图像效果,且满足实时性要求。上述改进后模型与实际摄像系统原理基本一致,增强了仿真系统视觉效果逼真性与沉浸感。由于该方法在工程上极易实现,可以在虚拟现实系统中广泛应用。

Abstract：

A way that could be improved was brought forward on the pin-hole camera model in allusion to the problems of the insufficient cue of depth. The relation formulas of the magnification and aspect angle were concluded according to the principles of digital zoom and the zooming function of visual channels was realized. The implementation of this method is simple and the effect is realistic. Besides,a simulation method for the course of adjusting focus was proposed,which brought blurry images into focus by particle system. Both vivid images and real time requirement could be achieved. The ameliorated model runs in the same way as the real camera system on the theory,thus enhancing the fidelity and immersion of visual effects. It is so easy to implement in project that this approach can be used universally in VR systems.     

An Optimization Model of Joint Tactics Organization Based on Chance-constrained Programming

Generally, the traditional troops optimization model adopts the method of certain mathematic programming. However, it is hard to satisfy the actual request of the accurately quantitative analysis with a large number of stochastic factors in modern high technique war. In this paper, the general characteristic of this problem is analyzed fully. A new method of establishing the optimization model of joint tactics organization based on chance-constrained programming is setted up. In the end, an example is demonstrated and solved by the genetic algorithm based on stochastic simulation.     

New data association technique based on ACO with directional information considered

Due to the advantages of ant colony optimization （ACO） in solving complex problems, a new data association algorithm based on ACO in a cluttered environment called DACDA is proposed. In the proposed method, the concept for tour and the length of tour are redefined. Additionally, the directional information is incorporated into the proposed method because it is one of the most important factors that affects the performance of data association. Kalman filter is employed to estimate target states. Computer simulation results show that the proposed method could carry out data association in an acceptable CPU time, and the correct data association rate is higher than that obtained by the data association （DA） algorithm not combined with directional information.     

有一定物理基础的核素浓度预测神经网络模型

为解决突发核电事故后使用机理模型预测放射性液态流出物迁移扩散,需长时间迭代计算的不足,提出了一种新型混合神经网络模型,该模型耦合了描述液态流出物在受纳水体中迁移扩散的组分输运方程和神经网络模型,采用并行多种群混合进化粒子群算法计算神经网络权值与阈值.论文以湖北咸宁大畈核电站受纳水体富水水库为研究对象,对事故工况下长半衰期核素迁移扩散进行模拟预测,研究结果表明有一定物理基础的神经网络模型是一种有效、可行的预测模型,预测结果与机理模型的模拟输出拟合度较好,新模型较传统的黑箱神经网络模型以及基于单调型先验知识的神经网络模型具有更强的泛化性能改善能力.     

人造目标SAR图像实时仿真

根据SAR图像特点,提出了一种利用图形加速的实时SAR仿真方法。方法采用散射中心模型,基于高频预估理论,用物理光学(PO)法与增量长度绕射系数(ILDC)法计算目标表面散射场强度分布,将强度值累积映射到SAR图像的分辨单元内,与成像系统传输函数卷积,获得近似仿真结果。散射、映射与卷积运算均借助图形加速,仿真可在数秒内完成。仿真结果与真实图像相比具有良好的视觉相似度。为算法测试而搭建的仿真平台OpenSARSim被作为开源项目,其所采用的部分技术可以应用于其它仿真领域。     

石油污染组分在包气带土壤中运移的数值仿真模型及应用

基于多孔介质流体力学和溶质动力学理论，建立了石油组分在土壤包气带内传输的数值仿真模型，初步探讨了石油组分苯和正戊烷在土壤中的迁移归宿及在水相和气相中质量分配，预测预报石油污染物泄漏后在包气带内迁移转化的动态及过程。仿真结果表明：苯和正戊烷在水相和气相中的浓度分布规律几乎相同，随着时间的延长，污染面积和范围逐渐增大，并且释放在水相、气相中正戊烷的质量大于苯的质量。这不仅对于定量化研究有机污染物在地下环境系统中迁移归宿，环境质量评价及污染预测、预报与污染防治提供科学的理论根据与途径，而且可为地下环境系统中石油组分的质量监测以及地下水资源的开发利用和科学管理提供技术支持。     

基于极限学习机的生化过程软测量建模

针对极限学习机方法隐层神经元数目过多的缺陷，提出一种改进的极限学习机方法。在单隐层前向神经网络的隐层中，增加一类分类神经元，从而形成了一种新的单隐层神经网络结构。针对不同类样本数不相同的问题，提出了处理方法，使得可以利用相同的隐层神经元对不同类的学习样本进行拟合，这使得网络的隐层神经元数目大大降低，从而简化了模型的结构，提高了神经网络的计算速度。将这一方法应用于诺西肽发酵过程，建立了菌体浓度的软测量模型，实现了菌体浓度的在线预估。     

矸石山渗滤液在地下水系统中运移的仿真分析

针对阜新矿区新邱南部矸石山对地下水污染实际情况,分析了矸石山渗滤液在含水层中运移的规律,建立了渗滤液在地下水系统中运移的耦合动力学仿真模型,利用该模型对渗滤液在地下水中迁移行为进行仿真分析,预测了污染物浓度的分布范围及随时空变化趋势。并与于家沟地区实际检测点数据进行对比,结果表明:渗滤液浓度拟合误差小于6%,从而验证了模型的可靠性和实用性,为地下水资源开发过程中污染控制与治理措施提供依据和理论指导。     

基于PO的进近着陆系统信道环境分析和仿真预测

针对机场环境日益复杂,进近着陆系统对于机场环境的影响十分敏感的实际问题,提出了一种基于物理光学理论(PO)的进近着陆系统信道环境分析和仿真预测方法。应用基于PO的信道散射环境分析方法,建立了系统的信号源模型和考虑信道环境影响的系统航道偏移模型,对受多散射体影响的信道环境下的进近着陆系统进行建模,结合实际问题,应用建立的模型对散射环境中的进近着陆进行了仿真,结果表明,基于PO的仿真预测方法能够较准确、合理地给出散射环境对系统的影响程度,具有预测机场配置物体对系统影响的能力。     

催化剂粉尘浓度软测量建模研究与应用

采用基于Mie理论的激光散射法测量催化剂粉尘浓度时,催化剂粉尘浓度与监测参数——入射光强、散射光强、出射光强以及烟气流量之间存在着复杂的非线性关系,给粉尘浓度的准确测梁带来困难。利用支持向量机优良的非线性映射和强大的泛化能力,建立了一个基于最小二乘支持向量机的催化剂粉尘浓度软测量模型,给出了相应的系统结构和算法,并通过网格搜索和交叉验证的方法对支持向量机进行参数选择。采用遗忘因子法和数据滑动时间窗技术对工业软测量模型进行在线校正,克服了工况条件发生改变时的估计偏差,提高了估计精度。仿真和实际运行结果表明基于LS-SVM的软测量模型具有较高的估算精度与泛化能力,为催化剂粉尘浓度的在线测量提供了一种简单、可靠的新方法。     

基于暂态挖掘的非线性时间序列故障预报

针对实际工程系统故障建模困难、现有故障预报方法实时性差的问题.从一类挖掘的角度,设计了一种基于一类支持向量机的时间序列暂态挖掘算法,提出了一种既不需要系统近似模型也不需要故障训练数据和先验知识的直接故障预报方法.在系统运行的同时实现学习和预报.提高了实时性.同时该方法简单易用,克服了传统方法在预报故障前需要预测系统未来状态的缺点.具有很强的应用意义.以釜式反应嚣为对象进行的仿真实验证明了方法的有效性.     

基于神经网络的辐射源识别系统设计

针对当前辐射源识别系统中存在的问题,提出了一种结合神经网络技术的辐射源识别新方法。该方法可以快速高效的识别各类辐射源,既有基于统计分析的辐射源识别系统的快速性,又有基于专家系统的辐射源识别系统的自适应性和准确性。实际仿真结果表明该方法是有效的,尤其对于参数不全、参数畸变的雷达辐射源,其识别率和识别置信度都有较大提高。在本文方法的基础上,设计出一种结构简单、快速有效的辐射源识别系统,具有一定的推广价值。     

氨糖发酵过程建模与工艺参数优化研究

针对目前生物传感器价格昂贵且检测精度低使得在氨糖发酵过程中难以获得准确实时的生物参数的现状,建立了最小二乘支持向量机模型以实现菌体浓度、产物浓度、底物浓度的预测。为了提高预测模型的精度,采用基于Levy飞行的改进多元宇宙算法对最小二乘支持向量机模型的若干参数进行优化。在此模型的基础上,以发酵完成时刻产物浓度最大为目标,通过改进的多元宇宙优化算法对发酵工艺参数进行了优化。仿真实验表明该方法取得了较高建模精度,提高了发酵最终产物浓度。