Aiming at different height of dam and different ratio of elastic modulus of the dam and dam foundation, this paper made a lot of calculation. Some conclusions are obtained. For the given global energy error limit, the calculation results can meet the required error to use adaptive mesh strategy of ANSYS program, and the influence of controlling parameters of mesh size can be ignored, and the stress singularity of dam heel doesn't influence the convergence of adaptive calculation.
然后将自适应有限元法应用于大坝有限元应力取值的研究，通过对不同坝高、弹模比的自适应计算，得到如下结论：(1)对于给定的全域能量误差限，采用ANSYS程序的网格剖分自适应策略，可以满足误差要求，并且基本不受单元尺寸改变控制因子的影响，不因坝踵部位应力集中而产生不收敛现象。
We also show that this merit function can provide a global error bound for SCCP with the uniform Cartesian P-property.
第三章我们主要感兴趣的是求解SCCP的几种可能的算法中的正则光滑牛顿算法。
Query requests a semantic analyzing module and a global query processing module to process the case of homograph and thesaurus. A Sub-query localization processing module processes case caused by different measure unit. Redundancy discarding module processes case caused by error of redundancy data.
查询请求语义分析和全局查询阶段处理同形异义和异形同义情况、子查询本地化模块处理计量单位不同所造成的影响、冗余处理阶段根据策略处理有冗余结果集中内存在错误数据所造成的冲突。
A 300-Year Error-----Three-Hundred-Year Devil's DEN on NB of SNB, Two Decrypting Equation Books of 《 The Laws (ISBN1442131241)》 and 《 The Laws Plus (ISBN144213285X)》; A Three-Hundred-Year Error`s Decryption, The Sword of the Third Signorino—Nostradamus` Sword of Great Seven Number that Makes the Global 6.5 Billions People Be Not Any More Buried Alive by the Earthquake in the World; Ali Baba and the Forty Thieves Open Simsim to the Sun and Moon Treasure House of the God to Get the Heaven`s Nuclear Cimelia: Nuclear Les Trois Mousquetaires and Big Three`s Triumvirate Carry out Three Nuclear Controls and Monitors to Make the Global 6.5 Billions People No Longer Lie Under Big Deathful Nuclear Radiation Here and There
全球拉响核警报：核泄露、核块熔化，全球15%以上核弹参与，同时大面积污染土壤和水资源，全球80%以上的人遭受严重核放射性照射辐射一个300年的错误；两本解密方程书《 The Laws (ISBN1442131241)》和《 The Laws Plus (ISBN144213285X)》；三百年错误的纠正令全世界65亿人从此不用再被地震无情活埋和被海啸无望吞噬；核三剑客令全世界65亿人从此不用再遭受致命核辐射
Epipolar geometry constraint is calculated by nonlinear error propagation at initial time to accelerate the speed of stereo matching; global and local methods are integrated to extract foreground objects precisely; and only pixels on objects'contour are reprojected into 3D space to fasten the depth calculation process.
使用视觉计算实现视频增强现实遮挡处理，用非线性误差扩散模型提高了对极几何的求解精度；将全局和局部高斯核相结合，提高并加快了物体前景轮廓线的提取精度和速度；仅用轮廓线上的点进行立体匹配的方法加快了真实物体深度反求的速度。
In this paper, by using some characteristic quantities such as value function, trust region radius, etc. in trust region method, we present a global error bound for the distance between a feasible solution and the optimal solution under the condition that the gradient vector of the objective function is strongly monotone.
本文利用信赖域方法中的几个特征量（由预测下降量给出的价值函数与信赖域半径等），在目标函数的梯度向量是强单调的条件下，为约束最优化问题的可行解与最优解之间的距离提供了一个全局误差界。
The accuracy of Global Position System is affected by several kinds of error sources. Nowadays, the tropospheric and ionospheric delays are the main error sources in the GPS positioning.
全球定位系统定位精度受许多误差的影响，目前最主要的误差来源为对流层误差和电离层误差，其中电离层误差对於RTK GPS 定位具有决定性的影响。
The algorithm is used to find the global minimum. A useful expression is obtained for the approximate error of the networks.
算法可以求得全局最优点，得到反映网络误差的一个有用的表达式。
For the sequence-style training, it is showed that there always exist N local minimum points in the error surface. For the batch-style training, it is showed the sufficient and necessary condition that FNNs can obtain global minimum solutions with null cost is that the range space R of the outer-supervised signal matrix Y must be included in the range space R of the hidden output matrix X, i. e., R? R. In addition, it is also showed that the sufficient condition that the FNN can obtain global minimum solutions with null cost is M≥N.
对于序贯训练方式，我们证明在其对应误差表面上总是存在N个局部最小点；对于批处理训练方式，我们证明网络获得零代价全局最小解的充要条件是，外监督信号矩阵构成的列空间必须位于模式样本在隐层张开的矩阵的列空间内；而网络获得零代价全局最小解的充分条件是，隐节点数M大于或等于非吻合的样本数N。
As an application of the new results for Pcharacteristic, we also give global error bounds for the EVLCP under P characteristic.
作为P特征的新等价结果的应用，我们给出具有P特征的扩展的垂直线性互补问题的全局误差界结果。
The data of satellites position, DR sensors and milemeter is processed by Kalman filters. Then the estimated output values of vehicle position, the direction and position error are inputed to map matching module, and the coordinates of matching location are calculated. After global optimum estimation, the final combined position output are given.
卡尔曼滤波器处理卫星定位和DR传感器及里程仪的数据，再将车辆位置、行车方向和定位误差等滤波估计值输入地图匹配模块，计算出匹配位置坐标，进行全局最优估计，得出最终的组合定位输出。
Settings Contents Iterations Maximum number of iterations to be performed Convergence Convergence criterion value; if the maximum absolute change in all the link flows between consecutive iterations is less than this value, convergence is achieved and the assignment procedure stops Alpha The global default value of the parameter in the BPR function Beta The global default value of parameter in the BPR function Function Error term distribution function for stochastic user equilibrium Error Percentage error for the error term used in stochastic user equilibrium assignment
收敛：收敛判别标准值；经过连续迭代，如果路段交通量变化值的最大值小于此值，那就认为收敛了，并且终止迭代； 3。
That`s less than 1% of the company`s global income -- a rounding error.
这在谷歌全球收入中不到1%，是可以忽略不计的。
Meanwhile, based on the principle of neural network terminal attractor, a thorough study on a terminal attractor BP learning algorithm and its application is conducted by means of nonlinear dynamics theory, consequently a new adaptive attractor BP algorithm to determine the appropriate number of hidden units in multilayer feed-forward neural network is first presented, it can adjust adaptively the number of hidden units in the training process of synaptic weights, and guarantee to reach the global minimum of the error measure function, as well converge faster than the conventional BP algorithm.
同时，根据非线性动力学理论，在神经网络终极吸引子原理的基础上，研究了神经网络终极吸引子算法及其应用，首次提出隐单元数自适应吸引子AHBP算法，使网络能够在学习过程中自适应调整隐单元数，确保网络权值学习收敛到全局极小点。而且收敛速度比普通BP算法快。以国产SZC-654型三坐标测量机为例进行了十八项基本运动误差神经网络建模的研究。
Using Hong(2001)'s method based on Cross Covariance Function and Error Correction Model, we study Granger causality and information spillover s among major global crude oil markets including London, New York, Dubai as well as Tapis and Minas in Southeast Asia.
采用基于协相关函数的Hong(2001)方法和基于协整理论的误差修正模型系统研究了全球原油市场的Granger因果关系与信息溢出。
Using Hong(2001)'s method based on Cross Covariance Function and Error Correction Model, we study Granger causality and information spillovers among major global crude oil markets including London, New York, Dubai as well as Tapis and Minas in Southeast Asia.
采用基于协相关函数的Hong(200l)方法和基于协整理论的误差修正模型系统研究了全球原油市场的Granger因果关系与信息溢出。
The architecture employs fuzzy systems to approximate nonlinear function which concludes fault function, uses fuzzy sliding controller to compensate for the fuzzy approximation error, through altering the structure of controller compensate for the influence due to actuators fault, Global asymptotic stability is established in the Lyapunov sense, with the tracking errors converging to a neighborhood of zero.
利用模糊逻辑系统来逼近未知函数包括故障项，用滑模控制补偿模糊逼近误差，通过改变控制器结构来补偿执行器故障所带来的影响，同时基于Lyapunov方法进行了稳定行分析。通过仿真证明该方法的有效性。
AFEM based on global error estimate rebuild mesh according to refinement factor. As a result, global error is distributed on every element and adaptive optimization method could be used on element.
基于全局误差估计的自适应有限元法根据单元细化因子对网格进行重新划分，从而使整个声场空间的误差分担到各个单元上，对单元进行自适应优化处理。
Mainly introduced the global localization technology and the geographic information system technology, introduced the localization principle and the position error analysis;
主要介绍了全球定位技术和地理信息系统技术，介绍了定位原理和定位误差的分析；
For a class of nonlinear time-varying repetitive systems with time delay、initial deviation、output error and condition disturbances, in arbitrary initial condition, the PD type closed-loop iterative learning tracking control problem which satisfies the global Lipschitz conditions is investigated.
针对一类存在初始偏差、输出误差和状态扰动的时滞非线性时变重复系统，在任意初值条件下，研究了系统满足全局Lipschitz条件下的PD型闭环迭代学习跟踪控制问题。