界量

Using the method adopted by Wangcheng, we have obtained the N-S equations with flexion as fundamental quantity in finite domain.

本文采用王诚提出的方法，得到了有界区域内的以弯曲度为基本量的N-S方程。

Ward start the flame, the top magic shield, feathers are hanging, reading goats, sheep被灭continue to run 30 yards or so, reading a few lines Austrian red shield to offset some of the sheep again, slow down, read the two Austrian red, spring screen, Austrian red shells and then by means of screen attack, fire red hair do not, the minister will use a shield, Prayer, recovery, and other means to disperse back to their own blood, if the drive time out slow heart perfusion burst open like Shenhuo, came as far as possible range, I chose the first round of the outbreak is a rapid-fire, the ice ring, the strong Austrian missile barrage, gas will be the order of silence during this period depends on啥时候pay the rhythm of the fighting and the opponent's blood, and good luck capable of seconds, seconds, then the minister can not read should be treated, because the silence had to pay, this time I chose to hang a slowdown, flash out, disappear, eat and drink immediately, in the pastor you into combat before I can eat 4,5 At that time Master of the blood should be very little anti-shield, and the minister should also be tight blue, my way is to drag a silent shield up to prevent interrupt Shadow Magic, rub-bing ring side of silence, and then wake-up call, time over 8 seconds, then the victory in sight.

开局火焰结界，顶级魔法盾，羽毛都挂上，读羊，被灭继续羊，跑到30码左右，读个奥冲抵消一些真言盾再羊，减速，读两个奥冲，弹幕，然后以奥冲弹幕等手段攻击，火冲吹风不用，牧师会用盾，祷言，恢复，和驱散等手段给自己回血，如果驱掉减速开灌注读心爆神火之类的，尽量闪出射程，我选择的第一轮爆发是出速射了，冰环，开奥强飞弹，弹幕，气定这个顺序，这期间啥时候交沉默要看战斗的节奏和对手的血量了，运气好了能秒，秒不了的话牧师应该会读治疗了，因为沉默交过了，这时候我选择挂个减速，闪现出去，消失，马上吃喝，在牧师把你拉进战斗前能吃4，5口，这时候法师的血也应该被盾反弄得没多少了，而牧师的蓝也应该吃紧，我的方法是拖到下一个沉默，撑盾防止暗影魔打断，擦边冰环沉默，然后唤醒，8秒读满，则胜利在望。

The dissertation focuses on performance gradient estimation and SRO algorithm of DES aswell as other related topics,such as modeling of DES and discrete event simulation,performancemeasure Hessian matrix estimation,Stochastic Approximationand so on.General Semi-Markov Processis adapted to describe behaviour of DES and discrete eventsimulation.A general result is obtained and applied to analyze strong consistency of performancegradient Perturbation Analysisestimator.Conditions for effectiveness of InfinitesimalPerturbation Analysisare also discussed.New PA algorithms,such as server PA,PA usingDirac generalized function,job sequence PA etc.,are proposed to deal with discreteparameters.Furthermore,strategies of combining PA algorithms with certain iterative rules areanalyzed widely in order to optimize DES parameters and some SRO algorithms are constructed toslove DES optimization problems.

论文将DES性能分析领域中性能梯度估计和单运行仿真优化作为主要的研究对象，探讨了与性能梯度估计有关的DES和离散事件仿真过程建模、高阶导数估计以及随机逼近等理论和方法；建立了具有一般性的DES和离散事件仿真的广义半Markov过程模型；获得了分析性能梯度估计量强相合性的理论结果和无穷小摄动分析算法应用的条件；提出了诸如服务台数目的摄动分析、奇异摄动分析、串联生产线和加工车间中工件加工序的摄动分析等一系列摄动的思想和算法，对连续参数和离散参数都进行了深入的研究；对摄动分析算法同优化规则结合实现仿真优化的策略也进行了广泛的分析，构造了新型的变界截尾的随机逼近算法，并用于DES的单运行仿真优化研究；最后，给出了DES性能分析方法的实证性研究结果。

By Llibre's algebraic invariant theory, at first on the basis of classification for quadric and quadric homogeneous binary polynomial, the plane linear and cubic homogeneous polynomial differential systems are classified.

利用Llibre的代数不变式理论，首先由二元二次和二元四次多项式的分类结果，对一次和三次齐次多项式微分系统进行代数分类，同时补充了已有结果中出现的漏洞；其次，由称共变张量空间的性质，对缺二次项的三次微分系统在保证轨线走向不变的前提下进行代数分类，使分类后同类系统的示性多项式有相同零点；最后通过讨论一类简单系统的有界性说明了分类的方便方处。

The results indicate that the flow stress of the alloy increases with increasing strain rate and decreasing deformation temperature. The flow stress increases with increasing strain until the stress reaches the peak value, then the flow stress remains constant, which indicates that dynamic recrystallization happens during deformation. The flow behaviors are described by the hyperbolic sine constitutive equation, and the activation energy calculated is 337.75 kJ/mol. The as-forged microstructure consists of refined α2/γ and γ grains, and the grains are much homogeneous than before. The B2 phase distributes uniformly at the grain boundary of α2/γ and γ grains. The B2 phase decreases with increasing deformation temperature.

结果表明：流变应力随着应变速率提高和变形温度降低而增大；在变形过程中，流变应力随着变形量增大而增大，当流变应力达到峰值后趋于平稳，表明合金在变形过程中发生了动态再结晶；热变形过程的流变应力可采用双曲正弦本构关系来描述，平均激活能为337.75 kJ/mol；从合金的组织演化过程中可以看出，合金中不均匀的原始组织得到明显均匀化，变形后的组织是由α2/γ层片晶团和γ晶粒组成的双态组织，在α2/γ层片晶团和γ晶粒的晶界交界处发现分布均匀的B2相，并且随着变形温度升高B2相数量逐渐减少。

Results Factors affecting senile dementia in proper order of intensiveness were advanced age,history of stroke,smoking,hypertension,nature and family members living together,excluding the difference of positive results caused by the effect of cultural level.It seemed that the other factors were not very important in the Logistic multivariate regression analysis.

结果在排除文化程度本身对痴呆筛查量表阳性界值的差异性后，对影响老年期痴呆的最大因素依次为：高龄、脑卒中史、吸烟、高血压、性格和家庭同住人口类型等，而其他因素在多元回归分析中似乎所占密度不大。

Due to the loss of O and Si at reducing atmosphere, the amount of intergranular phase is lower than that of the sintering additives.

研究发现，由于烧结助剂中的O和Si在烧结过程还原气氛下的损失，使得烧结体中晶界相的含量明显低于烧结助剂的加入量。

In addition, an effective method is proposed to select the β-reducts. First, we calculate a precision parameter value to obtain the subsets of information system that are based on the least upper bound of the data misclassification error. Next, we measure the quality of classification and remove redundant attributes from each subset.

本论文也提出一个根据精确参数来选择最简化集合的方法，这方法首先利用资料错误率的最小上界来决定精确参数值，并利用所获得的精确参数值来寻找资讯系统的子集合；接著计算每一个子集合的分类品质并利用分类品质的量测移除子集合中多余的属性，而删除多余属性的子集合即β的最简化属性集合。

In the proposed algorithm, we make use of the Lipschitz condition to determine lower bounds of functions over each rectangle.

在算法中，我们运用Lipschitz条件来确定目标函数和约束函数的在每个n-矩形上的上下界，对于n-矩形的分割，我们采用选择珩矩形最长边的二分法，同时我们采用了一些矩形删除技术，在不大幅增加计算量的前提下，起到了加速算法收敛的效果。

Normal synovial membrane, OA synovial membrane and normal cancellous bone were obtained as negative or positive control.

结果 界膜组织中PDGF-Bm RNA的表达量由高到低依次为OA滑膜、溶骨区界膜、正常松质骨、非溶骨区界膜、正常滑膜(P＜0.01)。

But we don't care about Battlegrounds.

但我们并不在乎沙场中的显露。

Ah! don't mention it, the butcher's shop is a horror.

啊！不用提了。提到肉，真是糟透了。