A variational formulation based contact algorithm for rigid boundaries in two-dimensional SPH applications

Smooth particle Hydrodynamics (SPH) is one of the most effective meshless techniques used in computational mechanics. SPH approximations are simple and allow greater flexibility in various engineering applications. However, modelling of particle-boundary interactions in SPH computations has always been considered an aspect that requires further research. A number of techniques have been developed to model particle-boundary interactions in SPH and allied methods. In this paper, an innovative approach is introduced to handle the contact between Lagrangian SPH particles and rigid solid boundaries. The formulation of boundary contact forces are derived based on a variational formulation, thus directly ensuring the conservativeness of the governing equations. In addition, the new elegant boundary contact force terms maintain the simplicity of the SPH governing equations.     

An alternative contact/impact identification algorithm for 2d structural problems

This paper presents a new algorithm to identify the occurrence of impact (contact) among structures. Structures are modeled here by finite elements and the resulting formulation is employed to solve two-dimensional frame impact problems by the Lagrange Multiplier technique. The proposed algorithm is based on potential theory, largely employed in various fields of physics. The potential theory is used together with integral equations making possible to compare relative positions among nodes of different structures involved in a collision process. The present technique may be used to model the contact prevision among two or three-dimensional structures or between structures and rigid obstacles. In order to show the applicability of the proposed technique, numerical answers for 2D problems are compared with both analytical solutions and numerical results given by other authors.     

BEM and FEM analysis of Signorini contact problems with friction

In this paper, we present a comparative study of the boundary element method (BEM) and the finite element method (FEM) for analysis of Signorini contact problems in elastostatics with Coulomb's friction law. Particularities of each method and comparison with the penalty method are discussed. Numerical examples are included to demonstrate the present formulations and to highlight its performance.     

A Direct analysis of elastic contact using super elements

Solutions to contact problems are important in mechanical as well as in civil engineering, and even for the most simple problems there is still a need for research results. In the present paper we suggest an alternative finite element procedure and by examples show the need for more knowledge related to the compliance of contact surfaces. The most simple solutions are named Hertz solutions from 1882, and we use some of these solutions for comparison with our finite element results. As a function of the total contact force we find the size of the contact area, the distribution of the contact pressure, and the contact compliance. In models of finite size the compliance depends on the flexibility of the total model, including the boundary condition of the model, and therefore disagreement with the locally based analytical models is expected and found. With computational contact mechanics we can solve more advanced contact problems and treat models that are closer to physical reality. The finite element method is widely used and solutions are obtained by incrementation and/or iteration for these non-linear problems with unknown boundary conditions. Still with these advanced tools the solution is difficult because of extreme sensitivity. Here we present a direct analysis of elastic contact without incrementation and iteration, and the procedure is based on a finite element super element technique. This means that the contacting bodies can be analyzed independently, and are only coupled through a direct analysis with low order super element stiffness matrices. The examples of the present paper are restricted to axisymmetric problems with isotropic, elastic materials and excluding friction. Direct extensions to cases of non-isotropy, including laminates, and to plane and general 3D models are possible.     

舰船水下接触爆炸的SPH算法研究

舰船水下接触爆炸问题涉及多相物质耦合,而且密度分布极不均匀,阻抗严重不匹配,大变形,强冲击等因素都使传统数值算法很难进行数值模拟。本文旨在通过改进传统SPH算法,并将其应用于水下接触爆炸问题：提出了变光滑长度的链表搜索算法,提高了计算效率,也保证了计算精度;总结了SPH方法对密度极大不均匀问题处理思想;研究了SPH方法的前、后期处理技术,提高了对问题的处理能力,增加了计算结果的可视性。文中最后成功模拟了水下接触爆炸问题,结果与理论一致,验证了所提出的理论和方法的有效性和可行性。另外,分析了不同粒子间距比对数值模拟的影响,结果表明,当粒子的间距比小于2:1时数值结果相对稳定,等于5:1时数值计算崩溃。     

The dependence of penetration velocity on impact velocity

Recent experimental measurements show that eroding long-rod penetration velocity is a linear function of impact velocity over a very wide range of impact velocities and for an interesting range of rod–target material combinations. These experiments all show that U=a+bV, where U and V are the penetration and impact velocity, respectively, and “a” and “b” are constants for given projectile and target materials. Numerical simulations also show that U=a+bV. The accumulation of these results suggests that a linear relationship between penetration and impact velocity may be fundamental over a very large range of impact velocities. A linear relationship between penetration and impact velocity has a number of implications. Some implications of this result for the Tate–Alekseevskii model are briefly examined in this paper.     

运用改进的遗传算法进行框架结构损伤检测

运用一种改进的遗传算法来进行结构的损伤检测研究。该方法在传统遗传算法的变异算子里引入了一种被称为零变异率因子的参数，使得种群中时刻保持一定数量的零值元素，即相当于用结构的损伤只是发生在局部这个信息约束了传统的遗传算法，从而使得检测的结果更加准确。通过对某一框架模型试验数据的损伤检测研究，证明了该方法应用于实际工程结构的可行性，同时阐明了实际应用中的若干重要问题。     

提高多普勒计程仪性能的矩阵算法

文中介绍了一种精确的矩阵算法,从而使多普勒计程仪的性能有很大程度的提高。借助于高精度频率测量、载体姿态的实时输入、最小二乘方估计以及矩阵算法,从而使载体速度矢量3个分量的测量精度优于0.1%。借助于高精度时延测量、最小二乘方估计以及矩阵算法,使载体下方的局部海底参数得以建立,从而使载体高度的测量精度有望优于1%,并且能获得助航海底姿态。     

Physics of deformation and fracture at impact loading and penetration

The structure, dislocation substructure, and mechanical properties of the targets made of four aluminum alloys after a impact loading by kinetic energy projectile have been investigated. The formula for approximation of the ballistic limit velocity by indentation technique is proposed. It has been shown that the maximum nonequiaxiality of the grain shape, increase of dislocation density, and decrease of dislocation cell size correspond to the 40–70% of plastic deformation at static compression for the investigated aluminum alloys.     

一种基于改进MCMC算法的模型修正方法

标准马尔可夫链蒙特卡罗(MCMC)算法不易收敛、拒绝率高,使其应用受到限制。在贝叶斯方法中引入最大熵值法来估计参数的后验概率密度函数最大值,进而将布谷鸟算法中新鸟巢更新的思想融入Metropolis-Hasting(MH)抽样算法得到改进的MH抽样算法,同时使用支持向量机(SVM)建立待修正参数与有限元模型输出之间的代理模型,以提高模型修正的计算效率。分别使用三自由度线性系统和平面桁架模型来验证本文方法的有效性,结果表明:修正后样本的马尔可夫链混合性能好,停滞概率低,修正后参数相对误差均小于2%。     

An examination of deviations from hydrodynamic penetration theory

The hydrodynamic theory of penetration (HTP) was first developed in the U.S. during WWII, and independently and essentially simultaneously in England. Since then the theory has proved very useful in understanding and predicting results of many penetration experiments. The assumptions and limitations of HTP were well stated in the initial paper. The most obvious limitation is that, strictly speaking, HTP only applies to hydrodynamic materials, i.e., both the projectile and the target have no strength. But for nearly all cases of interest, penetration does depend on material strengths, even at quite high velocities. Consequently, effects of projectile and target strength on penetration physics have been studied by many researchers, and modified versions of HTP have been proposed. While material strength is an important reason for deviations from HTP, it is not the only one. Other assumptions underlying HTP are steady-state behavior and incompressibility. In this paper we present new numerical simulation results that examine and quantify deviations from HTP due to compressibility for several material combinations of interest as a function of impact velocity. For these calculations all the materials are modeled as having zero strength. This is done in order to separate effects of compressibility from effects due to material strength. Some discussion of transient effects is also provided.     

改进遗传单纯形混合算法在光纤对接中的应用

本文首先给出了纤端光场分布的3D模型并针对单模光纤与激光器的对接分析了错位、角度偏差等对接效率影响因素。通过与传统的爬山法进行比较,提出将遗传单纯形混合算法应用到光纤对接中。单纯形法是一种局部搜索方法,这种寻优方法收敛速度快,它与遗传算法组合起来可以改善单独使用遗传算法收敛速度慢的缺陷。同时,针对光纤对接要求初始种群小所造成的个体多样性差的缺点,改进了单纯形的反射、压缩系数,有效地避免了遗传操作的过早收敛,并通过仿真结果证明了混合算法可以在10代之内找到光纤对接的最佳耦合点。     

改进的EKF算法在目标跟踪中的运用

过程噪声和测量噪声影响Kalman滤波的性能,通常很难得到它们准确的值。提出观测噪声和过程噪声实时估计的自适应算法。该算法可以用在非线性和机动目标跟踪问题中,不必预先知道准确的噪声方差。重新估测观测噪声方差矩阵,可以较好地消除由观测噪声带来的误差;建立一个简单的线性Kalman滤波器对过程噪声进行实时估计,这对于机动目标来说是必要的,因为原有的过程噪声将受到加速度影响,不能包含全部的信息。实验表明,该算法保证EKF稳定性,提高了跟踪性能。模拟实验300次后,X,Y方向位置均方误差分别为7.8099,9.6838。     

PCA-BP算法在地面爆破振动中的应用

为了更加准确地预测地面爆破的质点峰值振动速度,提出应用一种PCA-BP算法,该算法首先利用主成分分析对爆心距、高程差、总药量、炮孔深度、单段最大药量等地面爆破振动影响因素进行研究,然后结合BP神经网络算法对其爆破质点峰值振动速度进行预测。结果显示:利用PCA-BP算法的预测结果更接近工程实测值,平均相对误差为7.748%,远小于用传统萨道夫斯基经验公式进行预测的平均相对误差32.654%,说明将PCA-BP算法应用到爆破振动工作中是比较可行的,对评估地面振动危害有一定的指导意义。     

改进的人工免疫算法在图像配准参数优化中的应用

借鉴禁忌搜索的思想改进了人工免疫网络算法(aiNet),提出一种禁忌人工免疫网络算法(TS-aiNet).在算法中引入禁忌表,禁忌在网络迭代中亲和力不再增加的细胞,通过特赦准则赦免一些被禁忌的优良状态;增加记忆表,保存成熟的记忆细胞;重新定义高斯变异方式,保证多样化的搜索.利用Markov链分析了该算法的全局收敛性,通过对典型系统的仿真实验分析了该算法的性能,并与克隆选择算法和opt-aiNet算法进行了比较,最终将改进的算法运用到红外与可见光图像配准中,像素级配准精度可以达到0.5像素.实验结果表明,该算法在多模态搜索空间中具有更好的全局收敛性、稳定性和发现极值点能力,能够克服早熟现象,提高图像配准的速度和精度,是一种有效的全局优化方法.     

基于SPH方法对不同药型罩线性聚能射流形成及后效侵彻过程的模拟

为了解决传统基于网格的数值方法在模拟线性聚能射流问题时因大变形而导致网格畸变使计算难以进行的问题,本文通过自编程实现的光滑粒子法（SPH）对不同药型罩线性聚能装药射流形成及其侵彻金属靶板的过程开展了数值模拟研究,所实现的算法可以为线性聚能射流数值模拟研究提供新途径。本文所开展的研究首先基于已有的线性聚能射流试验模型进行模拟分析,采用SPH方法有效实现了线性聚能射流的形成过程,数值模拟获得的射流头部速度与试验比对误差在10%以内。然后建立了装药质量、药型罩质量和装药横截面宽度相同的前提下不同药型罩线性聚能射流模型,数值模拟获得不同药型罩形成的射流特征以及侵彻金属靶板的开口宽度和侵彻深度随时间的变化规律。研究得到的不同药型罩线性聚能射流形成及后效侵彻规律可为线性聚能射流的设计提供参考。     

Parallel contact algorithms for nonlinear implicit transient analysis

This paper presents a nonlinear implicit transient formulation for parallel contact analysis. A new lumping algorithm based on the penalty method is developed to enforce contact constraints. This algorithm has the effect of incorporating contact generated connectivity and eliminating ill conditioning in the system stiffness matrix caused by the use of a large penalty parameter. Communication schemes are also developed to facilitate contact searching and enforcement in a parallel environment. Numerical analyses were conducted to verify the accuracy of the proposed algorithm. In addition, several contact simulations are also performed to study the performance of this new algorithm under different circumstances. Efficiency plots are also presented for one of the simulations to evaluate the performance of this parallel implicit contact formulation.This paper is based upon work supported by, or in part by, the U.S. Army Research Office under grant number DAAD19-99-1-0235. The authors also acknowledge support under a DoD HPC Challenge Project.     

SELP声码器基音周期参数量化合成改进算法

针对正弦激励线性预测(SELP)声码器在解码端使用整型基音周期会引入额外舍入误差的问题,提出了一种码本整型化修正算法,该算法通过对码本中的各个码字在线性域进行取整,达到与解码端反量化过程一致的效果。测试结果表明,该算法能够有效地降低基音周期对数域矢量量化的量化误差,并且能够将合成语音的平均意见得分(MOS)提高0.05左右,明显改善合成语音质量。     

采用改进的基于余量的算法实现人脸识别中最佳Gabor特征的选择

基于2DGabor变换的人脸特征描述已经受到了很多人的关注。然而现有的Gabor特征维数较高,而且具有冗余性,因此选择最佳的Gabor特征用于人脸识别显得尤为的重要。利用最大余量原理的特征选择算法在目前的机器学习研究中已经占据了重要的地位。本文在基于余量的迭代搜索法(Simba)的基础上,引入了一种新的选择算法:基于余量的共轭梯度法(Cgmba),它只需较少次迭代就可以找到最佳解。我们在IMM人脸库上进行了实验,实验结果表明:尽管只使用了一半不到的特征,但Cgmba和Simba的识别率却分别提高了3.75和1.25个百分点,同时也证实了我们提出的Cgmba明显优于Simba。最后我们对Cgmba选择的Gabor特征的分布情况进行了分析,可以看出较大尺度的特征相对于较小尺度的特征对于分辩人脸的细微差别具有同等的重要性,而且在垂直,135°方向的特征具有更强的分辩能力。