非定常输运基于香农熵的自动调整样本数策略

对于非定常输运问题提出了一种基于香农熵的自动调整样本数策略。将每一计算步的总样本数划分为若干批并逐步模拟每批中的粒子, 可以在每批粒子模拟结束后通过计算得到该时间步幸存粒子属性分布对应的香农熵值。采用在线收敛性诊断方法, 一旦通过香农熵值序列判断对应的幸存粒子属性分布已经收敛, 则可以提前结束本时间步的计算。对一个空间一维非定常输运模型的计算结果表明, 该策略可以显著减少每一计算步的实际样本数且保持最终的结果基本不变, 从而减少了计算时间, 提高了计算效率。     

多物理耦合计算中动态输运问题高效蒙特卡罗模拟方法

多物理耦合计算在众多领域都有重要应用.如果其包含粒子输运过程,用蒙特卡罗方法模拟粒子输运常占据大部分的计算时间,因此多物理耦合计算中动态输运问题的高效蒙特卡罗模拟方法意义重大,其不可避免地依赖于大规模并行.基于动态输运问题的特点,本文提出了两种新方法:一是针对输运燃耗耦合计算的新型计数规约算法;二是动态输运计算样本数自适应算法.两种算法都能在保持计算结果基本不变的前提下使计算时间大幅减少,从而提高了效率.     

强光电离团簇并行数值模拟

 为研究强激光电离氢原子团簇,在理论上采用1维氢原子团簇的经典粒子动力学模型,结合粒子对(PP)算法及粒子模拟（PIC）方法,采用自行搭建的9节点并行集群系统,利用消息传递接口（MPI)与OpenMP混合编程模型进行了并行数值模拟计算,获得了较为理想的计算加速比。并且引入了弛豫时间参数,有效地处理了粒子间的碰撞过程,在极大简化计算量的同时,保留了物理本质。所得模拟结果与已有的实验结果符合较好,表明该并行计算模型是稳定、可行的。     

基于多光谱吸收的交替迭代优化算法

基于模拟退火算法的吸收光谱技术可以实现高温气体温度浓度的二维分布重建。但其计算效率过低,不适合于大规模的场反演计算。针对于该问题,提出了一种基于多光谱吸收的温度场和浓度场重建的交替迭代优化算法。首先对浓度场的计算提出了线性化方案,将原系统半线性化,从而弱化了原系统的非线性;其次对待测温度场、浓度场提出了交替迭代方案,每一步迭代降低一半的未知变量个数;最后用优化技术引入惩罚项来克服重建过程的病态性,在每一步迭代时能自动调整正则化参数,保证了迭代解收敛于精确解。仿真和实验结果表明该算法具有较强的抗噪声能力,在保持重建精度的前提下,能够显著的缩短计算时间,适用于大规模的场反演计算。     

一种非定常N-S方程并行求解设计

为了解决计算流体力学(CFD)中非定常计算与越来越大的计算量,并行计算已成为一种现实有效的选择.论文首先研究了一种并行区域分解策略,该策略简单而高效,但需要算法配合.为此,采用了一种与并行完全兼容的隐式方法DP-LUR方法.通过双时间步长法,将DP-LUR方法延伸应用到非定常计算中而不改变其原有的性质.最后分析了并行编程中的主要难点,提出解决方法,即采用中间数据分离节点下标与处理,并给出了并行程序的总体结构.     

等概率符号化样本熵应用于脑电分析

样本熵(或近似熵)以信息增长率刻画时间序列的复杂性,能应用于短时序列,因而在生理信号分析中被广泛采用.然而,一方面由于传统样本熵采用与标准差线性相关的容限,使得熵值易受非平稳突变干扰的影响,另一方面传统样本熵还受序列概率分布的影响,从而导致其并非单纯反映序列的信息增长率.针对上述两个问题,将符号动力学与样本熵结合,提出等概率符号化样本熵方法,并对其物理意义、数学推导及参数选取都做了详细阐述.通过对噪声数据的仿真计算,验证了该方法的正确性及其区分不同强度时间相关的有效性.此方法应用于脑电信号分析的结果表明,在不对信号做人工伪迹去除的前提下,只需要1.25 s的脑电信号即可有效地区分出注意力集中和注意力发散两种状态.这进一步证明了该方法可很好地抵御非平稳突变干扰,能快速获得短时序列的潜在动力学特性,对脑电生物反馈技术具有很大的应用价值.     

并行模拟退火算法优化衍射光学元件设计

模拟退火(SA)是解决复杂非线性优化问题的好工具,开发其并行性也成为近些年的研究热点和方向。针对衍射光学元件(DOE)的束匀滑应用,结合消息传递接口(MPI)并行编程方法,开发并实现了具有良好收敛效果和并行效率的SA算法。在8个中央处理器(CPU)的小型并行计算平台上进行了入射口径310mm、出射20阶超高斯0.4mm×0.8mm聚焦光斑的优化设计,使用不到原来1/6的时间就能获得比原串行算法更好的匀滑效果,从而验证了该并行SA算法的快速有效性。     

流动和传热问题的分区并行计算

针对交错网格下的SIMPLE数值算法实施了分区并行计算方法,在小型局域网下实现了流动和传热问题的并行数值计算.对两个经典的流动和传热问题的数值模拟实验表明,所建立的并行计算环境和分区并行算法能够得到正确的和收敛的数值结果.但与串行计算结果相比,并行计算误差明显大于串行计算误差.对并行算法做出的性能分析表明,所给出的并行算法得到了明显的加速效率.随着计算规模的增大,加速比和并行效率提高更显著.     

一维双曲守恒律的龙格-库塔控制体积间断有限元方法

给出数值求解一维双曲守恒律方程的新方法——龙格-库塔控制体积间断有限元方法(RKCVDFEM),其中空间离散基于控制体积有限元方法,时间离散基于二阶TVB Runge-Kutta技术,有限元空间选取为分段线性函数空间.理论分析表明,格式具有总变差有界(TVB)的性质,而且空间和时间离散形式上具有二阶精度.数值算例表明,数值解收敛到熵解并且对光滑解的收敛阶是最优的,优于龙格-库塔间断Galerkin方法(RKDGM)的计算结果.     

蒙特卡罗临界计算全局计数问题新策略研究

基于一个裂变源分布对应香农熵序列的在线收敛性诊断方法,为提高全局计数整体效率的均匀裂变点算法将在首次激活迭代步和首次判断收敛迭代步的最大值之后启动.之后,整体精度指标将每隔固定迭代步数计算一次,一旦达到事先设定的精度标准,整个临界迭代计算将提前终止.这一判断过程将一直持续到事先设定的最大迭代步数为止.通过一个反应堆基准问题C5G7模型的计算表明上述新策略在合适的参数下有助于提高蒙特卡罗临界计算全局计数问题的整体效率.     

A residual-based message passing algorithm for constraint satisfaction problems

Message passing algorithms, whose iterative nature captures complicated interactions among interconnected variables in complex systems and extracts information from the fixed point of iterated messages, provide a powerful toolkit in tackling hard computational tasks in optimization, inference, and learning problems. In the context of constraint satisfaction problems (CSPs), when a control parameter (such as constraint density) is tuned, multiple threshold phenomena emerge, signaling fundamental structural transitions in their solution space. Finding solutions around these transition points is exceedingly challenging for algorithm design, where message passing algorithms suffer from a large message fluctuation far from convergence. Here we introduce a residual-based updating step into message passing algorithms, in which messages with large variation between consecutive steps are given high priority in the updating process. For the specific example of model RB (revised B), a typical prototype of random CSPs with growing domains, we show that our algorithm improves the convergence of message updating and increases the success probability in finding solutions around the satisfiability threshold with a low computational cost. Our approach to message passing algorithms should be of value for exploring their power in developing algorithms to find ground-state solutions and understand the detailed structure of solution space of hard optimization problems.     

Real-space parallel density matrix renormalization group with adaptive boundaries

We propose an improved real-space parallel strategy for the density matrix renormalization group (DMRG) method, where boundaries of separate regions are adaptively distributed during DMRG sweeps. Our scheme greatly improves the parallel efficiency with shorter waiting time between two adjacent tasks, compared with the original real-space parallel DMRG with fixed boundaries. We implement our new strategy based on the message passing interface (MPI), and dynamically control the number of kept states according to the truncation error in each DMRG step. We study the performance of the new parallel strategy by calculating the ground state of a spin-cluster chain and a quantum chemical Hamiltonian of the water molecule. The maximum parallel efficiencies for these two models are 91% and 76% in 4 nodes, which are much higher than the real-space parallel DMRG with fixed boundaries.     

Entropy of Non-equilibrium Stationary Measures of Boundary Driven TASEP

We examine the entropy of non-equilibrium stationary states of boundary driven totally asymmetric simple exclusion processes. As a consequence, we obtain that the Gibbs–Shannon entropy of the non equilibrium stationary state converges to the Gibbs–Shannon entropy of the local equilibrium state. Moreover, we prove that its fluctuations are Gaussian, except when the mean displacement of particles produced by the bulk dynamics agrees with the particle flux induced by the density reservoirs in the maximal phase regime.     

Application of entropies to the study of the decoherence of magnetopolaron in 0-D nanosystem

Many methods have been experimented to study decoherence in quantum dot (QD). Tsallis, Shannon and Gaussian entropy have been used to study decoherence separately; in this paper, we compared the results of the Gaussian, Shannon, and Tsallis entropies in 0-D nanosystem. The linear combination operator and the unitary transformation was used to derive the magnetopolaron spectrum that strongly interacts with the LO phonons in the presence of an electric field in the pseudoharmonic and delta quantum dot. Numerical results revealed for the quantum pseudo dot that: (i) the amplitude of Gauss entropy is greater than the amplitude of Tsallis entropy which in turn is greater than the amplitude of Shannon entropy. The Tsallis entropy is not more significant in nanosystem compared to Shannon and Gauss entropies, (ii) with an increase of the zero point, the dominance of the Gauss entropy on the Shannon entropy was observed on one hand and the dominance of the Shannon entropy on the Tsallis entropy on the other hand; this suggested that in nanosystem, Gauss entropy is more suitable in the evaluation of the average of information in the system, for the delta quantum dot it was observed that (iii) when the Gauss entropy is considered, a lot of information about the system is missed. The collapse revival phenomenon in Shannon entropy was observed in RbCl and GaAs delta quantum dot with the enhancement of delta parameter; with an increase in this parameter, the system in the case of CsI evolved coherently; with Shannon and Tsallis entropies, information in the system is faster and coherently exchanged; (iv) the Shannon entropy is more significant because its amplitude outweighs the others when the delta dimension length enhances. The Tsallis entropy involves as wave bundle; which oscillate periodically with an increase of the oscillation period when delta dimension length is improved.     

The Network Source Location Problem: Ground State Energy,Entropy and Effects of Freezing

Ground state entropy of the network source location problem is evaluated at both the replica symmetric level and one-step replica symmetry breaking level using the entropic cavity method. The regime that is a focus of this study, is closely related to the vertex cover problem with randomly quenched covered nodes. The resulting entropic message passing inspired decimation and reinforcement algorithms are used to identify the optimal location of sources in single instances of transportation networks. The conventional belief propagation without taking the entropic effect into account is also compared. We find that in the glassy phase the entropic message passing inspired decimation yields a lower ground state energy compared to the belief propagation without taking the entropic effect. Using the extremal optimization algorithm, we study the ground state energy and the fraction of frozen hubs, and extend the algorithm to collect statistics of the entropy. The theoretical results are compared with the extremal optimization results.     

Uncertainty relations expressed by Shannon-like entropies

Besides the well-known Shannon entropy, there is a set of Shannon-like entropies which have applications in statistical and quantum physics. These entropies are functions of certain parameters and converge toward Shannon entropy when these parameters approach the value 1. We describe briefly the most important Shannon-like entropies and present their graphical representations. Their graphs look almost identical, though by superimposing them it appears that they are distinct and characteristic of each Shannon-like entropy. We try to formulate the alternative entropic uncertainty relations by means of the Shannon-like entropies and show that all of them equally well express the uncertainty principle of quantum physics.     

Rate of Mutual Information Between Coarse-Grained Non-Markovian Variables

The problem of calculating the rate of mutual information between two coarse-grained variables that together specify a continuous time Markov process is addressed. As a main obstacle, the coarse-grained variables are in general non-Markovian, therefore, an expression for their Shannon entropy rates in terms of the stationary probability distribution is not known. A numerical method to estimate the Shannon entropy rate of continuous time hidden-Markov processes from a single time series is developed. With this method the rate of mutual information can be determined numerically. Moreover, an analytical upper bound on the rate of mutual information is calculated for a class of Markov processes for which the transition rates have a bipartite character. Our general results are illustrated with explicit calculations for four-state networks.     

Maxwell's demon and the entropy cost of information

We present an analysis of Szilard's one-molecule Maxwell's demon, including a detailed entropy accounting, that suggests a general theory of the entropy cost of information. It is shown that the entropy of the demon increases during the expansion step, due to the decoupling of the molecule from the measurement information. It is also shown that there is an entropy symmetry between the measurement and erasure steps, whereby the two steps additivelv share a constant entropy change, but the proportion that occurs during each of the two steps is arbitrary. Therefore the measurement step may be accompanied by an entropy increase, a decrease, or no change at all, and likewise for the erasure step. Generalizing beyond the demon, decorrelation between a physical system and information about that system always causes an entropy increase in the joint system comprised of both the original system and the information. Decorrelation causes a net entropy increase in the universe unless, as in the Szilard demon, the information is used to decrease entropy elsewhere before the correlation is lost. Thus, information is thermodynamically costly precisely to the extent that it is not used to obtain work from the measured system.     

库仑碰撞截面在等离子体粒子模拟中的应用

在等离子体粒子模拟中,TA模型和NanBu模型被广泛用于处理库仑碰撞,这两种模型要求每个时间步长内全部粒子参与计算。为了降低参与碰撞的粒子数,提高库仑碰撞的计算效率,提出了一种基于截面的库仑碰撞模拟方法,并给出了库仑碰撞概率的计算公式。采用该方法对不同温度不同密度电子气的弛豫过程进行模拟,分别对比了电子速度分布函数、电子温度以及电子x、y方向上的温度与电子温度之比的模拟值与理论值,验证了该方法的准确性。在相同的小时间步长上,该方法相比TA模型计算效率提升可达40%以上。对于较大的时间步长,该方法仍能得到与理论解近似的模拟结果,相比Nanbu模型,在相同的精度下可取更大的时间步长,计算效率也有所提升。研究表明,该方法同样适用于电子-离子碰撞。因此在提高库仑碰撞计算效率上,该方法具有碰撞粒子数少以及适用于大时间步长的优势。