SPH并行计算方案及其在自由表面流动模拟中的应用

针对光滑粒子动力学主要计算量是近邻粒子搜索这一特点,提出了一种基于粒子分解的光滑粒子动力学并行计算方案。利用该方案可以方便地将任意串行光滑粒子动力学代码并行计算,而且每一个时间步内的信息传递量只和粒子总数有关,而和粒子的分布无关,因而特别适合于自由表面流动等大变形问题的并行数值模拟。对一个粒子总数为40万的三维溃坝问题的模拟结果表明:此方案能达到的最大加速比约为16,这一结果可能比空间分解方案(不考虑动态负载均衡)更优。     

SPH数值模拟中固壁边界的一种处理方法

在光滑粒子流体动力学 (SPH)数值模拟中尝试了一种处理固壁边界的边界力方法 ,给出了一种新的边界力形式。利用SPH方法及边界力方法对水坝坍塌和涌波进入静止水塘这两个自由表面流动问题作了数值模拟。数值模拟结果表明 ,在SPH计算中使用本文所给边界力处理固壁是行之有效的     

基于迭代卡尔曼粒子滤波的目标跟踪算法研究

粒子滤波在基于图像序列的目标跟踪中获得了广泛应用．针对其计算量较大的问题,提出一种迭代卡尔曼粒子滤波算法,将非线性跟踪问题分解为线性子结构的全局状态空间模型和非线性子结构的局部状态空间模型,利用粒子滤波在卡尔曼滤波估计值的局部范围内搜索目标,逼近真实目标状态．将实验结果与粒子滤波进行比较,结果表明,迭代卡尔曼粒子滤波减少了粒子数,降低了计算量,能够对高机动目标进行实时稳定的跟踪．     

基于FEM/SPH算法的弹丸侵彻土体数值模拟

基于有限元法(FEM)和光滑粒子流体动力学(SPH)结合的算法,用数值模拟了钢质弹丸对钢筒约束土体的侵彻过程。基于ANSYS/LS-DYNA显式动力分析和LS-PrePost后处理软件,形象再现了钢质弹丸冲击作用下土体飞溅形成漏斗坑和直线通道的物理过程。侵彻深度计算结果与实验数据吻合较好,钢质弹丸的速度和加速度时程曲线图与理论分析一致,体现了FEM/SPH算法的精确性。研究结果表明FEM/SPH算法在模拟侵彻土体方面具有可行性和有效性。     

基于UKPF的雷达与红外传感器融合跟踪算法

针对异类传感器融合跟踪中状态估计与量测的非线性问题,分析了粒子滤波器在解决非线性和非高斯问题的优势,提出了一种基于unscented Kalman粒子滤波器(UKPF)的雷达与红外传感器融合跟踪算法,给出了算法的详细步骤.然后将该算法用于三维空间的目标跟踪上,仿真结果表明该算法比现有算法具有更高的跟踪精度.     

Unscented粒子滤波算法在合成孔径声纳组合导航中的应用

研究了基于捷联惯导(SINS)、多普勒声速剖面仪(ADCP)的水下组合导航方法。建立了基于四元数的SINS/ADCP误差模型。针对一般粒子滤波算法中存在的粒子退化问题,对Unscented粒子滤波算法进行了研究,并应用于SAS导航模型。算法采用Unscented卡尔曼滤波(UKF)引入了最新观测量来产生粒子滤波(PF)的建议密度分布,提高了状态估计的性能。仿真结果表明:在SAS复杂的运动情况下,UPF算法比传统粒子滤波算法具有更高的精度。     

基于粒子群优化的稀疏分解最优匹配原子搜索算法

信号的稀疏分解能得到信号的稀疏表示形式,便于进一步处理,但其计算非常复杂,是一个NP问题.粒子群优化是群体智能优化算法,算法简单,易于实现,且搜索效果好.把粒子群优化算法用于稀疏分解的最优匹配原子的搜索,能降低稀疏分解复杂度,同时减少稀疏分解的超完备字典对存储空间的占用,以提高用稀疏分解理论进行信号处理的计算效率,满足或接近实时性的要求.实验证明,此方法切实可行.     

冷喷涂粒子速度影响因素的数值模拟与分析

冷喷涂过程中,粒子速度直接影响冷喷涂材料改性效果.由于射流流场为超音速气固两相冲击射流,用实验方法研究粒子速度的影响因素很困难,为此通过数值模拟方法再现流场中粒子速度的变化情况.用Fluent软件以冷喷涂技术中超音速气固两相射流流场为研究对象,对于气固两相流的模拟,则采用颗粒轨道模型(欧拉-拉格朗日模型)获得粒子在射流流场的速度变化情况;分析粒径、材料、喷涂距离以及载气等因素对粒子速度的影响.进一步优化冷喷涂工艺方案,得到了粒子特性与粒子到达基板前最大速度之间的优化关系.     

多速率交互式多模型粒子滤波研究

由于粒子滤波的算法原理,其计算量很大.研究了针对机动目标的交互式多模型粒子滤波器(IMMPF)算法和多速率交互式多模型(MRIMM)算法,提出了多速率交互式多模型粒子滤波器(MRIMMPF)算法.该算法是在交互式多模型粒子滤波器(IMMPF)的基础上引入多速率技术,期望在保持IMMPF的性能同时能够减少因为粒子滤波带来的计算量;最后通过和一搬基于EKF的IMM算法、IMMPF算法的比较,验证了该算法的有效性.     

激光辐照下树脂基复合材料温度场MSPH算法的并行化

改进的光滑粒子方法在模拟激光对树脂基复合材料的辐照效应时具有明显优势,但串行计算通常难以满足需求。为了提高程序的运行效率,基于消息传递界面并行编程环境实现了串行程序的并行化。将个人电脑完成的串行计算结果与峰值5万亿次的计算机集群完成的并行计算结果进行了比较,并对并行程序的加速比和并行效率进行了测试。数值计算表明,并行计算结果与串行计算结果一致,且并行效果显著,说明基于消息传递界面的并行化是成功的。     

改进的接触算法及其在光滑粒子流体动力学中的应用

采用以黎曼解描述粒子间相互作用的接触算法,模拟一维情况下包含间断的流场。在弱间断的条件下,采用弱波近似的黎曼解来描述粒子间的相互作用;在强间断的条件下,则引入非迭代黎曼解来描述粒子间的相互作用。并引入Taylor展开思想,提高了接触算法在自由边界的计算精度。该算法无需引入人为粘性,程序结构简洁。数值计算表明,改进后的接触算法能很好地描述强间断,并有效改善了原始接触算法在自由面的计算缺陷。     

Characteristics structure analysis on debris cloud in the hypervelocity impact of disk projectile on thin plate

In this paper, the gauge points setting is introduced in the SPH simulation to analyze the debris cloud structure generated by the hypervelocity impact of disk projectile on thin plate. Compared with the experiments, more detailed information of the debris cloud structure can be classified from the numerical simulation. However, due to the solitary dispersion and overlap display of the particles in the SPH simulation, accurate comparison between numerical and experimental results is difficult to be performed. To track the velocity and spatial distribution of the particles in the debris cloud induced from disk and plate, gauge points are locally set in the single-layer profile in the SPH model. By analyzing the gauge points’ spatial coordinate and velocity, the location and velocity of characteristic points in the debris cloud are determined. The boundary of debris cloud is achieved, as well as the fragments distribution outside the main structure of debris cloud.     

一种基于DSP的雷达/电视视频图像压缩解压缩技术

在基于多媒体DSP处理器DM642并行处理平台上,提出一种雷达、电视视频图像压缩解压缩技术系统方案。采用多尺度小波分解技术和MPEG4压缩解压缩技术,分别对采样后的雷达视频、电视及红外视频数据压缩并分发到网络,由终端接收解压缩后融合实现多层多窗口综合显示。应用该方案可以实现舰载视频传感器和显控台以及武器系统的动态组合,提高装备的使用灵活性。     

Thermal decomposition of ammonium perchlorate catalyzed with CuO nanoparticles

Ammonium perchlorate (APC) is the most common oxidizer in use for solid rocket propulsion systems. However its initial thermal decomposition is an endothermic process that requires 102.5 J·g⁻¹. This manner involves high activation energy and could render high burning rate regime. This study reports on the sustainable fabrication of CuO nanoparticles as a novel catalyzing agent for APC oxidizer. Colloidal CuO nanoparticles with consistent product quality were fabricated by using hydrothermal processing. TEM micrographs demonstrated mono-dispersed particles of 15 nm particle size. XRD diffractogram demonstrated highly crystalline material. The synthesized colloidal CuO particles were effectively coated with APC particles via co-precipitation by using fast-crash solvent–antisolvent technique. The impact of copper oxide particles on APC thermal behavior has been investigated using DSC and TGA techniques. APC demonstrated an initial endothermic decomposition stage at 242 °C with subsequent two exothermic decomposition stages at 297.8 °C and 452.8 °C respectively. At 1 wt%, copper oxide offered decrease in initial endothermic decomposition stage by 30%. The main outcome of this study is that the two main exothermic decomposition peaks were merged into one single peak with an increase in total heat release by 53%. These novel features can inherit copper oxide particles unique catalyzing ability for advanced highly energetic systems.     

矢量多边形并行栅格化数据划分方法

针对多边形并行栅格化中的负载不均衡问题提出一种新的数据划分方法,主要包括:迭代计算划分线的位置,在每次迭代中保证分块间的计算量大致均衡,完成数据划分、实现负载均衡;提出基于二叉树的划分结果融合策略,以解决跨边界多边形的融合问题。在多核CPU环境下实现并行算法,选用多个典型土地利用现状数据集进行测试。结果表明:针对不同类型多边形数据集,所提方法较传统方法可获得更高的并行加速比和更好的负载均衡;针对大数据量数据集,以多边形节点数为度量标准可更精确地估算分块计算量,从而更好地实现负载均衡。     

An efficient SPH methodology for modelling mechanical characteristics of particulate composites

Particulate composites are one of the widely used materials in producing numerous state-of-the-art components in biomedical, automobile, aerospace including defence technology. Variety of modelling techniques have been adopted in the past to model mechanical behaviour of particulate composites. Due to their favourable properties, particle-based methods provide a convenient platform to model failure or fracture of these composites. Smooth particle hydrodynamics (SPH) is one of such methods which demonstrate excellent potential for modelling failure or fracture of particulate composites in a Lagrangian setting. One of the major challenges in using SPH method for modelling composite materials depends on accurate and efficient way to treat interface and boundary conditions. In this paper, a master-slave method based multi-freedom constraints is proposed to impose essential boundary conditions and interfacial displacement constraints in modelling mechanical behaviour of composite materials using SPH method. The proposed methodology enforces the above constraints more accurately and requires only smaller condition number for system stiffness matrix than the procedures based on typical penalty function approach. A minimum cut-off value-based error criteria is employed to improve the compu-tational efficiency of the proposed methodology. In addition, the proposed method is further enhanced by adopting a modified numerical interpolation scheme along the boundary to increase the accuracy and computational efficiency. The numerical examples demonstrate that the proposed master-slave approach yields better accuracy in enforcing displacement constraints and requires approximately the same computational time as that of penalty method.     

基于云进化的遗传粒子滤波算法

针对粒子滤波器存在的粒子贫乏问题,提出了一种基于云模型改进的遗传重采样方法。选择操作采用相隔一定代数进行随机采样的方式,防止选择压力过大导致粒子贫化;利用Y云发生器实现变异操作,根据粒子的观测概率自适应控制搜索范围,在现有粒子的附近搜索精良粒子,在提高粒子有效性的同时增加了粒子的多样性。仿真结果表明：改进后的算法有效地解决了粒子的贫乏问题,提高了滤波性能。