多变量的

In presented method, the correlated variables are transformed into uncorrelated variables equivalently, and the performance functions in the correlated variables space are transformed into those in the equivalent uncorrelated variables space, in which the reliability sensitivity of the system failure probability with respect to the distribution parameters of the uncorrelated variables can be solved by the moment method for the uncorrelated variables. Finally the equivalent relationship between the correlated variables and the uncorrelated variables, the chain rule for derivative are employed to obtain the reliability sensitivity of the system failure probability with respect to the means of the correlated variables.

在所提的方法中，首先将相关变量空间等效转化为不相关变量空间，并将系统多个失效模式的功能函数转化为不相关变量空间的功能函数，进而利用矩方法得到多模式系统失效概率对不相关变量空间基本分布参数的灵敏度，最后利用相关变量空间与不相关变量空间的等价关系及复合函数求导法则，求得多模式系统失效概率对相关变量空间中变量均值的灵敏度。

The dynamic coherence analysis based on multivariate autoregressive modeling was applied in the observation of synchronization between brain areas during cognitive processing.The result indicates that the synchronization of relative brain areas happens automatically despite the distractive stimulus during object perceiving,but the disturbed attention affects the synchronization of those areas involved in the judgement and voluntary movement as the way it affects the reaction time.

基于多变量自回归模型的动态相干性分析方法，被用于观察不同脑区在认知过程中的同步化，结果显示，人在感知和识别物体时，相关的脑区自动地发生了同步化的神经活动，这种同步化被证实是不受干扰刺激影响的，但被干扰的注意力确实影响了与判断和主动运动有关的脑区的同步化，而且这种影响的趋势与反应时间的分析结果一致。

Optimal algorithm of combined nonlinear Hopfield network appears powerful validity in solving nonlinear planning, which includes nonlinear objective function, linear constrains and high-demension of decision-making variants, for its ability of nonlinear parallel computation. It is prominent among optimal algorithms because of its function of simply implement with electrocircuit hardware. Genetic algorithm is expressly suitable for optimal calculation regarding the planning of massive, highly nonlinear, inconsecutively differentiable and multiobjective function as well as objective function without analytical expression. However, it inclines to prematurity, as well as its limitation in ability of partial optimal search. Introducing the optimal algorithm of niche genetic simulated annealing to standard Genetic algorithm, can therefore improve the full-scale or partial search ability of Genetic algorithm effectively. It has a far-flung perspective in the field of systemic planning of water pollution control.

组合式非线性Hopfield网络优化算法所具有的非线性大规模并行计算能力在求解具有非线性目标函数、线性约束条件及高维决策变量的非线性规划问题方面显示出了强大的生命力，它易于电路硬件实现的功能更是在优化算法中独树一帜；遗传算法采用概率搜索技术，不受目标函数与约束条件的限制，特别适合大规模、高度非线性的不连续可微的多峰目标函数及无解析表达式的目标函数的规划问题的优化计算，但其存在容易早熟、局部寻优能力较差等缺点，本文在标准遗传算法中引入小生境技术及模拟退火算法有效地改善遗传算法的全局和局部搜索性能，提高了全局最优解的寻优质量，小生境遗传退火模拟优化算法在水污染控制系统规划中的应用前景极为广阔。

Presents the design of longitudinal control for ASTOVL lift-fan aircraft, a multi-variable, strong-coupling, complicated nonlinear system to make the system nonlinear offset and decouple, and the method used for linear system in designing the system to gain expectant flying qualities and the combination of aerodynamic control surfaces with thrust vectors to delay the control surfaces getting into saturation, and the minimum N-norm generalized inverse theorem of weighted generalized inverse adopted to apportion the moments commands, which make the control surfaces gain appropriate deflection and the simulation results to verify the dynamic inverse control laws designed really make the system nonlinear offset and decouple; when one command is given, only this output will response, and others do not response; when all commands are given together, the responses will be just the linear superpose of the response of each command, therefore the operation of the pilot will be simplified greatly.

摘 要： ASTOVL升力风扇飞机是多变量、强耦合、复杂的非线性系统，采用非线性动态逆理论设计控制律以实现系统的非线性对消及解耦，然后，采用线性系统的设计方法对系统进行设计，以使系统获得期望的飞行品质。同时，考虑到ASTOVL升力风扇飞机是由气动舵面和推力矢量融合作用的，为了延缓各舵面进入饱和的时间，采用加权广义逆中的极小N-范数广义逆定理分配力矩指令，从而使各舵面达到适当的偏转角。并将所设计的控制律进行了仿真，结果验证了所设计的动态逆控制律使系统实现了非线性对消和解耦，单个指令作用时，只有该项响应，其余不受影响；所有指令作用时，其结果就等于单个指令作用时的线性叠加，这样使驾驶员的操纵大大简化而易于控制，从而大大降低了飞行事故率。

In addition to, we emphatically introduce the multilinear separation of variables, and we expound the general solving process about it, else I solve the variable separation solution of the system about DS correlating the process.

另外还重点介绍了多线性分离变量法，对多线性分离变量法的一般求解过程作了阐述，还结合步骤具体求解了DS系统的分离变量解。

The effective heat-transfer temperature difference of every effect of multi-effect evaporation subsystem and the preheat temperature rise Δt_(p,j of every level of multi-level preheat subsystem of the design model were distributed by equal heating area, equal heat-transfer temperature difference, and free proration methods.

设计型模型对多效蒸发子系统各效有效传热温差Δt_i的分配和多级预热子系统各级预热温升Δt_的分配分别采用等面积法、等温差法、自由配比法等进行分配并采用矩阵法结合迭代法进行设计模拟计算，其中自由配比法巧妙地将具有复杂约束条件的决策变量Δt_i和Δt_转化为取值0～1的中间决策变量τ_i、τ_，为Δt_i和Δt_的自由搜索进而为优化设计提供了一个简便的方法；在常规设计模拟计算的基础上，建立了复杂多效蒸发系统优化设计模型，该模型以系统年总费用最小为优化目标，以生蒸汽温度T_0、冷凝器中二次蒸汽温度T_K、蒸发温差配比系数τ_i和预热温升配比系数τ_为决策变量，采用遗传算法结合矩阵法求解。

The author constructed a linear schemes type Region-Project-Join-Normal-Form based on lossless join decomposition, and gave the form of RPJNF through the presentation of equivalence between semantic variable independence and syntax variable independence under finite precision domain. Some other new constraint data dependencies including partitioned multi-region existence dependency and single region existence dependency were also presented. A result of variables set under linear transformation was discussed.

文中给出并证明了多区域存在依赖满足的几条推理规则；提出并证明了线性序约束关系进行模式无损连接分解的充要条件是变量集满足变量独立和多区域存在依赖；设计了一种基于模式无损连接分解的区域投影连接范式；通过证明有限精度域上语义变量独立和语法变量独立的等价性，给出有限精度域上RPJ范式的表达形式。

Many other temporaries are used to hold the result of a long-winded bit of code for easy reference later. These kinds of variables should be set only once. That they are set more than once is a sign that they have more than one responsibility within the method. Any variable with more than one responsibility should be replaced with a temp for each responsibility. Using a temp for two different things is very confusing for the reader.

许多其它的临时变量则是暂存一段冗长代码的计算结果以便于后面对其引用，此类的临时变量只应该被赋值一次，如果他们被赋值多次那么说明他们在方法中担负着多种职责，任何担负着多种职责的变量都应该被单职责的临时变量代替，一个临时变量用于不同目的是会把读者搞糊涂的。

Coal load in ball tube of pulverizing system is affected by many factors, this paper analyzes these factors and their relations.

由于神经网络技术的发展，神经网络在非线性控制方面显示出强大的优势，同时近年来出现的逆系统方法对于多变量非线性系统的控制指出了一条可算是光明的道路，鉴于以上两项技术的优越性，本文提出对BMPS进行神经网络逆系统控制的方法，首先通过机理分析的方法建立了BMPS的非线性动态数学模型，并依此模型设计了神经网络逆系统控制方案，由于此方案中需要用到球磨机磨筒内负荷的反馈，但是磨内负荷是一个不能直接测量的量，而且其测量方法是多年来人们一直在努力的方向却又一直不能较好地解决，本文提出对其进行神经网络测量的思想和方法，这一方法的仿真试验和现场实测结果均表明其研发造价低廉的优越性和测量较为准确的实用性。

Throughout comparing the above estimating methods, we have the following results: the estimators of the moments of the errors does not depend on the random effects, and that of the random effects does not depend on the errors, and then the corresponding asymptotic variances are very simple and optimal; when the random effects are multivariate, we can not construct different estimating equations for the random effects and errors respectively, which results that the asymptotic covariances of estimation are very complex and then the estimating efficiency is bad.

比较上述两种估计法，我们发现：当随机效应是一维的时侯，误差的各阶矩的估计不依赖不可观测的随机效应，随机效应的估计也不依赖误差，因此，估计的渐近方差结构特别简单也是最优的；而当随机效应是多维的，因为随机效应的协变量的影响，我们没有办法针对随机效应和误差的各阶矩分别建立估计方程，这导致所得的估计的渐近方差或者协方差矩阵特别复杂，从而估计的效果不是很好。

I am accused of being overreligious," she said in her quiet, frank manner,"but that does not prevent me thinking the children very cruel who obstinately commit such suicide.""

客人们在卡罗利娜·埃凯家里，举止就文雅一些，因为卡罗利娜的母亲治家很严厉。

Designed by French fashion house Herm è s, this elegant uniform was manufactured in our home, Hong Kong, and was the first without a hat.

由著名品牌 Herm è s 设计，这件高贵的制服是香港本土制造，是我们第一套不配帽子的制服。