查询词典 model

Accordingly, this paper proposes a short-term traffic flow prediction model of data fusion algorithm based on Filtering step by step. Compared with traditional prediction model through Kalman Filtering Theory, this model is not only reduce the computation and lower model to center processor's request, but also ensure estimate accuracy.

基于此，本文提出了一种基于分步式滤波的数据融合算法(Fusion Algorithm of Filtering Step By Step ，简记FAFSS)的短时交通流量预测模型，该模型与传统的基于卡尔曼滤波理论的预测模型相比，不仅减少了计算量，降低了模型对中心处理器的要求，而且保证了预测的精确度。

Word-table structure and structural model of COMS can be found in this paper, including text model、database model and FDB model.

给出了COMS的词表结构和结构模型，这些结构模型包括：文本模型、数据库模型和FDB模型。

This model contains two parts, one is conceptive model and the other one is reasoning model. The fuzzy reasoning algorithm of traffic control and the automatic detection algorithm for traffic congestion are integrated in the reasoning model.

设计的自动决策支持模型中包含了概念模型和推理模型，推理模型中集成了交通拥挤自动检测算法和拥挤交通量控制方案选择的模糊推理算法。

In cross-section analysis, there exists conditional convergence in China and spatial error model is better than spatial lag model. In panel data analysis, fixed effect spatial error model is better than fixed effect spatial lag model.

在横断面分析方面，中国大陆地级城市间存有条件收敛，以空间误差模型较佳；Panel Data分析则以固定效果空间误差模型优於固定效果空间落迟模型。

The traditional HMM for Web information extraction is sensitive to the initial model parameters and easy to lead to a sub-optimal model in practice. A hybrid conditional model to combine maximum entropy and maximum entropy Markov model is put formard for Web information extraction.

针对传统Web信息抽取的隐马尔可夫模型对初值十分敏感和在实际训练中极易得到局部最优模型参数，提出了一种最大熵和最大熵马尔可夫模型相结合的条件模型。

In this thesis, I summarize and implement the up-to-date language models, such as N-Gram Language Model, Interpolation N-gram Language Model, Context-Free Grammar and Probabilistic Context-Free Grammar, and establish the run-length language model Character-level and word-level run-length language model are established on the base of the concept of limit entropy and discrete stable information source and by using maximum likelihood between character strings.

在充分研究和比较了中文和西文特点的基础上，提出了变长汉语语言模型。应用极限熵、离散平稳信源和字符串之间的最大似然度等概念和方法，建立了字一级的变长语言模型和词一级的变长语言模型。实验的结果显示这种语言模型不仅比Bigram具有更低的复杂度，而且包含了更长距离内的信息，运行时间也大大小于N-Gram。

The redesigning frame makes up of two models: one is redesigning time decision model, which is based on continuous approximation model; the other is specific redesigning model, which comes from the model of location-routing problem.

文中用树编码遗传算法来求解该设计框架中的离散模型，并对遗传算法中的编码、解码、交叉、变异、免疫等各种算子做了详细的介绍。

The application of the lattice Boltzmann method in unsteady open channel flows was studied in detail in the paper. An existing lattice BGK model was mended; For the problem of negative local equilibrium and evolutionary distribution, we proposed a lattice Boltzmann model based on cell-population equilibrium, which is a direct non-negative approximation to the continuous Maxwell distribution. The model reduces the transport and collision, two basic evolution steps in the LB model, to transport of the non-equilibrium distribution.

本文系统的研究了格子Boltzmann 方法在明渠非恒定流中的应用问题，改进了一个现有的LB 模型，数值实验表明此改进是有效的；在平衡态分布函数及分布函数的非负性问题上，作者提出了一个基于单元均衡的格子Boltzmann 模型，通过直接对Maxwell 分布函数的离散化实现对速度分布函数的非负离散，这种方法将传统模型演化过程中的碰撞和输运两个步骤直接表现为粒子的输运。

Firstly, this paper investigated roundly the basic theory of the ambient vibration method, analyzed systematically the main methods of ambient vibration identification, and discussed their advantages and disadvantages. Then we established the finite element model of structural dynamics and test model and analyzed their mutual relation, improved MUSIC method and correlation function method in the harmonic identification, and gave the subspace method and correlation function method based on continuous model, which is multi-harmonic retrieve technique. And then, this paper also puts forward least-squares method to estimate directly structural physical and modal parameters; this approach replaced the derivative operator in the continuous time model with an equivalent approximation, and estimated the physical parameters of structures by the least-squares method. And then, we discussed the approximation methods of derivative and their convergence. Finally, this paper researched the modal parameters of a frame structure by the simulation and test. The results demonstrated that subspace method, correlation function method and the least-squares method are effective and stable respectively.

全面地研究环境振动识别方法的基本理论，系统地分析环境振动识别方法的主要方法，讨论它们的优点和不足；建立结(来源：6666AB44C论文网www.abclunwen.com)构动力学的有限元模型和试验模型，分析有限元模型与试验模型之间的相互关系；改进谐波识别中的MUSIC识别方法和相关函数识别方法，给出大型结构模态识别的连续型的子空间分解识别方法和相关函数识别方法，这个方法是多元谐波识别方法；提出直接识别结构的物理参数和模态参数的最小二乘方法，这个方法利用导数的近似值代替导数值，并运用最小二乘方法识别结构的物理参数，讨论导数近似值的选择方法，分析算法的收敛性；通过仿真计算和模型实验，研究高层框架结构的模态参数识别，验证子空间分解识别方法、相关函数识别方法和最小二乘识别方法的有效性，算法的稳定性；从理论和试验两个方面分析和研究噪声对识别精度的影响，影响识别精度的因素，以及部分模态成分的模态参数不能够被识别的原因。

First,the convex hull points of the characteristic region such as eyes,mouth etc. are confirmed,and,the integral contour points are obtained by fitting the curve.Then,the coordinate of the contour and the corresponding points of the adapted facial model are transformed to a new coordinate.At last,the model is adjusted in the new coordinate to get a good adaptation between the human facial model and human face image by revising the corresponding detail data in the model.

首先，对检测出的眼睛、嘴巴等特征区域，确定出其外凸包点集，通过曲线拟合得出其完整外轮廓，然后对此轮廓点以及已经粗调的对应模型点进行坐标转换，在转换后的坐标系中，对模型中相应的细节数据进行修改，实现模型和人脸图像的配准。

The split between the two groups can hardly be papered over.

这两个团体间的分歧难以掩饰。

This approach not only encourages a greater number of responses, but minimizes the likelihood of stale groupthink.

这种做法不仅鼓励了更多的反应，而且减少跟风的可能性。