Solving liveness problem for marked nets by exhaustive coverability trees

It is well known that a coverability tree of a Petri net cannot solve reaehability and liveness problems of the net because using symbol ω (infinity.) may lose some information. A solution to this problem is presented for a special kind of Petri net, marked net. With the combination of o and the increasing/decreasing information of token number, a new kind of coverability tree of marked nets, called exhaustive coverability tree (ECT), is proposed. It is shown with an example that an ECT can be used to detect deadlock.     

一类Poisson型分布的特征

证明了满足 EX²<∞ 的具有非退化分布的母体 X 服从 Poisson 型分布 P (λ,α, 1)的充要条件是 T₂- T₁ 关于 T₁有常回归,其中 T₁= X =1n
ni= 1X_i , T₂=
1n- 1
ni= 1( X_i-
X )²分别为子样均值和子样方差.     

关于Petri网归约问题

Petri 同归约问题是当今 petri 网研究领域的一个重要课题,本文从 Petri 网静态结构的角度出发,给出了三个层次的 Petri 网归约方法,并用代数的方法严格地证明了这些归约方法保持网的两大重要性质:活性和有界性。它们为 Petri 网的归约和分析的复杂性之间提供了一个灵活折衷的手段,能够对一个庞大、复杂的 Petri网进行不同程度的归约,以适应不同的需要。文中所给出的归约算法是基于网的关联矩阵的,具有高效,便于实现的特点。此外,本文还就多弧 Petri 网的归约问题和 Petri 网的分解问题作了简要讨论。最后还给出了哲学家用餐问题的 Petri 网新模型并对之进行归约。     

碳纤维增强复合材料编织网混凝土梁试验研究

将碳纤维增强复合材料（CFRP）编织成正交网格,并经环氧树脂胶固化形成受力骨架,代替钢筋混凝土梁中的钢筋,以解决普通钢筋混凝土中钢筋易锈蚀的问题。通过对3根不同配筋率的CFRP编织网混凝土梁的弯曲性能试验,观察在各级荷载作用下构件的应变变化、挠度发展、破坏过程及破坏特征,对其受弯承载力及抗弯刚度等弯曲性能进行分析。试验结果表明：CFRP编织网与混凝土有良好的粘结性能,二者协同受力性能良好;破坏前,试件有明显的塑性变形,有明显的破坏预兆,延性性能较好;随着配筋率的提升,构件承载能力、抗弯刚度较大幅度提升。试件破坏时,CFRP编织网纵向受力筋均未达到其抗拉强度,故在以后的试验中可对编织网的纵向受力筋施加一定的预应力,以提高CFRP编织网的利用率。     

描述与求解哲学家就餐问题的Petri网模型研究

研究了用Petri网描述操作系统中经典的“并发操作”的例子——哲学家就餐问题，并构造了不同的Petri网模型解决死锁和无限等待问题，最后讨论了轮流进餐时不同进餐次数情况下的Petri网构造问题。     

Topology, vibration and NMR spectra of fullerenes

The topological structure of the fullerenes with symmetryI _h, I;T _d, T_h, T;D _nd,D _nh, D_n(n = 3, 5, 6) has been investigated. The infrared and Raman active vibration modes and NMR spectrum of fullerenes are obtained by use of the combination of graph and group theory methods.     

扩展Petri网在PPCES建模中的应用

利用生产计划与控制专家系统(PPCES)进行启发式搜索,根据Petri网建模原理及PPCES的推理需要,对生产系统建模,定义了一类扩展有色、时间Petri网,用来解决Job-shop型生产系统中NP难度的调度问题.实践证明,这种方法不仅可以在增强模型描述能力的同时有效地降低其复杂度,还可以应用专家系统领域的知识指导系统进行冲突消解,从而避免了常规分析方法所带来的组合爆炸问题.     

神经网络训练的分解重组算法

本文介绍了把线性不可分问题分解为一系列线性可分子问题、对线性不可分问题进行求解的网络分解重组算法．还证明了该算法的收敛性．实例研究表明：该算法不仅可以得到神经网络的隐层空间目标和隐层单元数，而且提高了对线性不可分问题的求解速度，因此是一个非常有效的神经网络训练算法．     

Structural, magnetic and transport properties of Sc-doped La0.7Sr0.3MnO3

The Sc-doped La_0.7Sr_0.3MnO₃ polycrystallines were prepared by thermal decomposition of complexes of metal ions with ethylenediamine-tetraacetic acid (EDTA). The dependence of structural, magnetic and transport properties on Sc doping was investigated. Moderate Sc doping level results in a sharp drop in Curie temperature, an increase in resistance and enhancement of magnetoresistance effect. At temperatures aboveT _MI,R(T) can be described asR(T) = R ₀exp(T ₀/T)^1/4, which indicates that conduction takes place by hopping between the localized states. The experimental data reveal that the change in magnetoresistance property is related to the structure.     

Effect of Ni addition on the glass-forming ability and soft-magnetic properties of FeNiBPNb metallic glasses

The effect of Ni addition on the glass-forming ability (GFA) and soft-magnetic properties of an (Fe1-xNix)75.5B14.5P7Nb3 (x=0-0.6) alloy system were investigated. We found that the addition of Ni was effective in allowing the alloy to approach a eutectic point as well as increasing the thermal stability of the supercooled liquid. By increasing the amount of Ni,the supercooled liquid region (ΔTx),the reduced glass transition temperature Trg (Tg/Tl) and the Y parameter [Tx/(Tg+Tl)] increased from 49 to 75 K,0...     

一种基于混合Petri网的批过程动态调度算法

针对批处理过程动态调度问题的特点，以赋色时间限制混合Petri网为工具，建立了批过程动态调度的混合Petri网模型，将批过程的调度问题映射技术为求Petri网中变迁的最优激发序列问题。在此模型基础上，根据动态规划基本原理，以资源的终极利用即最短生产周期作为调度目标，提出了一种基于赋色时间限制混合网模型求取变迁的最优激发序列的方法，并针对实际情况中出现的复杂的中间产品存贮策略，讨论了在各存贮策略约束下该模型中时间限制区间的确定方法，最后以炼钢－连铸生产系统的过程调度为例，对该算法进行仿真。多次仿真结果表明，当任务和设备的初始时刻或者过程的动态过程发生改变时，调度结果各不相同，从而验证了该算法的有效性。     

基于时延脉冲耦合神经网络的AOE-网问题求解算法

 在时延脉冲耦合神经网络(DPCNN-Delay Pulse Coupled Neural Network)的基础上,提出了mDPC-NN(modified DPCNN)模型.mDPCNN能方便地用于有向网的最大代价路径和最小代价路径的求解,并将其成功地运用其于求解AOE-网(Activity On Edge)有关问题.该方法通过快速并行计算,一次正向计算即可求解AOE-网大部分问题,与传统方法相比体现了较好的优势.     

柔性制造系统Petri网模型的化简

在一类S3PR网柔性制造系统Petri网模型N中,位置特殊资源不可能出现在网的严格极小信标和基本信标之中.因此,对于包含此类资源的网系统,为了缩小网规模,需要对其化简.化简算法对网模型进行处理,将位置特殊资源及相关操作库所、变迁和弧删去.同原网系统相比,最终得到的网系统具有较少的库所、变迁和较简单的网结构.     

数学神经网络（Ⅲ）——关于泛函联接神经网络

从异或问题引出泛函联接网络，然后研究泛函联接网络的构造及其数学本质，指出所谓泛函联接均可由神经元的激发函数来实现，最后，关于学习算法提出一点注记，证明该系列论第２部分提出的学习算法具有一般性或通用性。     

The toxic effect of solubilizing excipients on <Emphasis Type="Italic">Tetrahymena thermophila</Emphasis> BF<Subscript>5</Subscript> growth investigated by microcalorimetry

The toxic effect of different solubilizing excipients on the growth and metabolism of Tetrahymena thermophila BF₅ (T.t.BF₅) at various concentrations was investigated by microcalorimetry. The thermogenic curves of T.t.BF₅ growth were determined at 28°C, and were evaluated by dynamic parameters. The results indicated that the values of growth rate constant (k), maximum power (P₁, P₂), peak time (T₁, T₂) and total quantity of heat (Q) varied for different excipients. There was a good linear relationship between k and concentrations (r>0.95, P<0.01). 5% inhibition concentration (IC₅) of poloxamer 188, Tween 80, PEG 600, PEG 400 and Tween 20 was 2.18, 1.07, 1.35, 0.58, and 0.045 mg/mL, respectively. After the principal component analysis (PCA), Q, k and P₁ could characterize the effect of these excipients on T.t.BF5 growth. Comprehensive evaluation indicated that compared with the control group, poloxamer 188 had the weakest toxicity and Tween 20 had the strongest toxicity.     

Signal语言对扩展型Petri网的分析

为了解决现有Ｐｅｔｒｉ网分析工具（可达性树和矩阵方程）不能很好对扩展型Ｐｅｔｒｉ网进行分析的问题，提出用Ｓｉｇｎａｌ语言对扩展型Ｐｅｔｒｉ网进行分析的方法．Ｓｉｇ－ｎａｌ语言不仅能够解答扩展型Ｐｅｔｒｉ网安全性、有界性等问题，还能够直观地表示扩展型Ｐｅｔｒｉ网中各变量之间的约束关系及特性，更好地揭示网络的动态变化行为．     

离散事件系统的受控Petri网模型中的冲突控制

本文研究了以受控Ｐｅｔｒｉ网为模型的一类离散事件系统的冲突控制问题，提出了一种综合冲突控制的方法．该方法首先把冲突问题转化为禁止状态问题，然后通过解决禁止状态的控制问题实现受控Ｐｅｔｒｉ网中的冲突控制．文中还给出了求解控制的有效算法．     

Novel sintering behavior of polystyrene nano-latex particles in filming process

Filming process of polystyrene nano-latex (NPS) particles was studied by a combination of various methods. For a constant annealing time of 1 h, the AFM images showed that the deformation and sintering temperatures for NPS particles were ca. 90℃ and 100℃ respectively. In spin-lattice relaxation measurements of solid state NMR, it is found that T_1L, T_1S and I_1L/I₀ increased significantly after annealing at 90℃ and above. DSC results showed that there was an exothermic peak near T_g after annealing for 1 h at the selected temperatures below 95℃; otherwise, the exothermic peak disappeared after annealing at 100℃ or above. The apparent density of NPS increased suddenly in the temperature range. The results implied that the macromolecules in NPS particles are in a confined state with higher conformational energy and less cohensional interactions which are the drive force for the sintering at a lower temperature compared with the multichain PS particles and the bulk polymer.     

“三软”复合顶板均压支护技术研究

王村煤矿属于"三软"复合顶板,巷道支护是一个长久探索的课题。为了解决这一问题,在13503工作面进行高强高预应力让均压锚网支护实验,并采用FLAC2D软件数值模拟优化让压均压支护系统,实践表明该支护系统能有效控制"三软"复合顶板巷道围岩变形问题,确实提高支护质量。该技术对解决本矿及类似地质条件的矿井巷道支护难题有重要的指导和借鉴意义。     

Aboveground biomass and corresponding carbon sequestration ability of four major forest types in south China

We estimated aboveground biomass carbon (TABC) and net carbon accumulation rates (TNEP) for trees in four major forest types based on national forest inventory data collected in 1994-1998 and 1999-2003. The four types were Pinus massoniana forest, Cunninghamia lanceolata forest, hard broad-leaved evergreen forest and soft broad-leaved evergreen forest. We analyzed variations in TABC and TNEP for five stand ages (initiation, young, medium, mature and old). In both time periods, estimated TABC in all four forest types increased consistently with forest stand age and the oldest stage had the largest TABC compared with other stages. Broad-leaved forests (hard and soft) had higher TABC than needle-leaved forests (Pinus massoniana and Cunninghamia lanceolata) for each of the five age stages. The difference of TABC between broad-leaved and needle-leaved forests increased with forest stand age. Comparison of estimated TNEP by age category indicated TNEP increased from the initiation stage to the young stage, and then decreased from the mature stage to old stage in all four forest types. TNEP for any particular stage depended on the forest type; for instance, broad-leaved forests at both the mature and old stages had greater TNEP than in needle-leaved forests. A logistic curve was applied to fit the relationship between TABC and forest stand age. In each period, correlations in all four forest types were all statistically significant (P < 0.01) with R2 > 0.95. TABC was therefore predicted by these regression functions from 2000 to 2050 and the mean TNEP during the predicted period was estimated to be about 41.14, 31.53, 75.50 and 75.68 gCm-2a-1 in Pinus massoniana forest, Cunninghamia lanceolata forest, hard broad-leaved forest and soft broad-leaved forest, respectively. Results from both forest inventory and regression prediction suggest broad-leaved forests are greater carbon sinks, and hence have greater carbon sequestration ability especially in the mature and old stages when compared to needle-leaved forests. Broad-leaved forests should have high levels of carbon sequestration when compared with needle-leaved forests in south China.