子模型

And the tangent linear operator of the forward model also exists, which is the same as the resolvent of the tangent linear model derived by the conventional method.

对连续"开关"，代价函数的梯度处处存在的，非线性方程的切线性算子也是存在的，并且与传统方法得到的切线性模型的解算子是一致的。

Then we record the temperature of indirect monkshood moxibustion as the temperature boundary condition of the human model. The values of experiment match the numerical simulation ones.

并用红外热像仪记录艾炷隔附子饼燃烧过程的温度，将其作为边界条件施加在人体横截面模型上，模拟临床上穴位点施加隔附子灸的情况，实验和数值模拟数据吻合较好。

Two optimization schemes, which be used to constructed one-way operator in mixed domain, are described.

本文着重对混合域中单程波算子构造的两种优化方法进行分析，讨论他们与其他单程波算子构造方法之间相互关系，并用上述方法计算变速介质的脉冲响应及偏移Marmousi模型数据，以此说明该类方法的偏移精度。

Two optimization schemes, which be used to constructed one-way operator in mixed domain, are described. Firstly, their characters are analyzed, and the relations between optimization method and other methods of constructing one-way operator are given. Then In order to show precision of one-way operator constructed by two optimization methods, The impulses of varying velocity medium and migration results of Marmousi model are presented.

本文着重对混合域中单程波算子构造的两种优化方法进行分析，讨论他们与其他单程波算子构造方法之间相互关系，并用上述方法计算变速介质的脉冲响应及偏移Marmousi模型数据，以此说明该类方法的偏移精度。

During the iterations the trial step is truncated when necessary in order to keep the strict feasibility of the iteration points. It is shown that the LMTI approach and the STI approach have global convergence properties under some reasonable assumptions and will achieve local quadratic convergence rate under some even stronger conditions.

LMTI方法主要利用由有限内存BFGS公式的紧致形式获取的目标函数海色阵的近似矩阵构造信赖域问题模型并对其进行求解；而STI方法在构造的低维子空间上求信赖域子问题的解。

A hybrid model of short-term load forecasting based on chaotic theory, correlation and neural networks is presented in this paper. Firstly, reconstruct attractors in phase spaces using chaotic theory, Secondly fit the attractor's evolvement using BP neural networks, because selecting neural network's input training data using Euclid distance and correlation, improve neural network's associative memory and ratiocinative ability, can better fit the attractor's evolvement.

提出一种将混沌理论、关联度和神经网络相结合的短期负荷预测模型，首先利用混沌理论重构负荷时间序列的相空间吸引子，然后用BP 神经网络来拟合空间吸引子的演化，由于使用空间欧氏距离和关联度联合来选取神经网络的训练样本，这样就提高了神经网络对负荷序列混沌特性的联想和泛化推理能力，能够更好的拟合吸引子的演化。

Firstly, reconstruct attractors in phase spaces using chaotic theory,Secondly fit the attractor s evolvement using BP neural networks, because selecting neural network s input training data using Euclid distance and correlation, improve neural network s associative memory and ratiocinative ability, can better fit the attractor s evolvement.

提出一种将混沌理论、关联度和神经网络相结合的短期负荷预测模型，首先利用混沌理论重构负荷时间序列的相空间吸引子，然后用BP神经网络来拟合空间吸引子的演化，由于使用空间欧氏距离和关联度联合来选取神经网络的训练样本，这样就提高了神经网络对负荷序列混沌特性的联想和泛化推理能力，能够更好的拟合吸引子的演化。

Using the wave equation prestack depth migration on SEGEAGE 2_D salt dome model as an example, the twoorder generalized screen propagator increases 30% calculation work relative to the conventional generalized screen propagator.

以SEGEAGE二维盐丘模型数据的波动方程叠前深度偏移为例，二阶广义屏传播算子相对于常规广义屏传播算子增加了30%的计算量。

Comparing the unsteady flow fields obtained by CFD with the measurement results with SPIV, it can be found that the CFD results can capture the main characteristics of the flow fields observed in the experiment results, however, the precision of the CFD needs to be further improved, and to predict the detail flow structures, such as the tip leakage vortex in the compressor rotor, by CFD still has some difficulties, thus, some work should be done to improve the turbulence model for more accuracy prediction of the flow fields by CFD for the complex flow in turbomachinery.

定常计算结果和非定常计算结果都不能很好地反映出转子叶尖泄漏流，转子尾迹和泄漏流在静子通道中的演化，及其与静子的相互作用等复杂流动情况。开展叶轮机适用的湍流模型研究，对于提高叶轮机内部复杂流动数值模拟的准确性具有非常重要的价值。

Three typical terrains are simulated using wave superpose method based on the work of full exploration of terrain model and its realization Effect of wavelet number on simulation result is studied, the results indicate that satisfactory results can be obtained by selection of a certain wavelet number.

在充分探讨了地形模型和随机过程实现的基础上，采用波叠加法仿真了3种典型地形。研究了子波数对仿真结果的影响，说明选取一定数目的子波即可得到满意的结果。

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。