特征函数

The common method, that all strong-correlation terms of the model are eliminated, can bring the loss in the engineering application, so the new method is proposed that the identified model reserves some correlation. The augmented matrix A is constructed by the outputΔW and the matrix S. The"determinating order based on ratio of determinant"is brought out to screen the strong-correlation terms in the structure identification. The latent root estimation is improved in screening the eigenvalues and eigenvectors. Thus the estimation precision is improved greatly.The consistence check of guidance instrument error coefficients of flight test and ground test is the purpose of flight experiment. The causes of inconsistency of the two models are analyzed. The hypothesis test of linear regression model based on F statistics is proposed to check the consistence.Finally, the instability of error coefficients is probably caused by the change of the flight environments, therefore, the relation between the error coefficients and flight environment is analyzed. The approach is presented to identify SINS guidance instrument error models and compensate the error in the segmented sections corresponding to the change of vertical acceleration of aircraft.

在结构辨识中，常用的方法由于将模型中的强相关项全部剔除而给工程应用带来损失，因此，本文提出了新的有益思想，即在保留一定相关性的基础上进行辨识：将输出向量ΔW与环境函数矩阵S构成增广矩阵A，然后采用"比定阶行列式"来剔除相关向量的方法，这样既可以尽可能多地保留了对落点影响大的强相关参数，又可以对落点影响小的强相关参数给予剔除；在参数估计中，改进了特征根估计中特征根和特征向量的筛选方法，提出"近零"准则，从而大大提高了参数估计的精度；再者，鉴于天地模型"一致性"检验是飞行试验和SINS制导工具误差系数分离的主要目的，因此，本文又深入分析了造成天地模型不一致的原因，提出了采用基于F统计的线性回归模型假设检验方法来进行捷联制导工具误差模型的天地"一致性"检验；最后，鉴于飞行环境剧烈变化可能会对惯性仪表误差系数稳定性带来一定的影响，因此本文深入地分析了SINS制导工具误差系数与外界环境的关系，提出了基于过载变化大小的分段辨识和分段实时补偿的算法。

VQ is easy and tend to realize, HMM can comminute the speech dynamically, it can embody not only the dynamic characteristic but also the stationary characteristic of the speech. GMM is a special HMM, it uses combination of many weighted Gaussian density function to approach the distributing density of speaker characteristic vector in characteristic space, so it is also a good way.

VQ方法是最简单又易于实现，HMM方法能将语音动态分割，既能体现语音的动态特征，又能体现语音的平稳特性，GMM用多个加权高斯密度函数的组合来逼近说话人特征矢量在特征空间的分布密度，收到了较好的结果。

According to the characteristic of image sequence, we define a function of motion smoothness and provide the constant acceleration model of feature point.

针对非刚体图象序列的特点，定义了特征点运动的平滑代价函数，并提出了特征点运动的常加速度模型，在此基础之上，将非刚体特征点对应的建立问题表示为一个带约束的优化问题，并用Hopfield网络将其实现。

The experimental system extract the low-level features of images such as HSV histogram, the texture got from coexistence matrix, color correlogram, and according to the characteristic of our image database, design the evaluation function such as the average rank ratio to evaluate and compare the performance of different integration of different features including semantic, and validate the active effect of feedback using experiment results.

该原型系统提取了HSV直方图、共生矩阵纹理、颜色相关图等底层特征，同时根据图像数据库的特点，设计了平均位置比值等评价函数对语义特征与底层特征的各种结合的检索性能进行了比较分析，并对反馈效果进行了实验验证。

Firstly, Geodesic distance is introduced to overcome the weakness of Euclidean distance. Then, neighborhood rough set model is introduced to make full use of neighborhood local information and also can avoid the discrete process of numerical attributes.

最后将一个二进制DE算法运用到特征选择，为了设计合理的适应度函数，本文首先针对基于欧氏距离类内类间评价函数的不足，引入流形中的测地线距离，力求保持数据集内在特性的基础上，提出了一个基于测地距离的适应度函数。

A generalized Gaussian Laplacian eigenmap algorithm based on geodesic distance is proposed,which incorporates geodesic distance and generalized Gaussian function into the original Laplacian eigenmap algorithm.GGLE algorithm can adjust the similarities between nodes of neighborhood graph,and can preserve the different degrees of local properties by using super-Gaussian function,Gaussian function or sub-Gaussian function.Moreover,GGLE can avoid the deficiency of Euclidean distance by using geodesic distance when neighborhoods of data points are enlarged for preserving more neighborhood relations.Experimental results show that the global low-dimensional coordinates obtained by GGLE have different clustering properties and different degrees of preserving local neighborhood structures when different generalized Gaussian functions are used to measuring the similarities between high-dimensional data points.(3) An ensemble-based discriminant algorithm based on GGLE is proposed.

该算法将测地线距离和广义高斯函数融合到传统的拉普拉斯特征映射算法中，可以调整近邻图结点间的相似度，通过选择超高斯、高斯或者次高斯函数来实现不同程度的近邻局部特性的保持；而且当需要保持更多的近邻关系使得数据点邻域增大时，采用测地线距离可以避免欧氏距离度量不合理的缺陷；实验结果表明在用不同的广义高斯函数度量高维数据点间的相似度时，局部近邻结构保持的程度是不同的，GGLE获得的全局低维坐标也呈现出不同的聚类特性。

The relationship between the class of continuous functions satisfying Lipschitz condition and the class of polar-functions of generalized Brownian Sheet is obtained.

讨论了N指标d维广义Brownian Sheet极函数的特征，得到了满足Lipschitz条件的连续函数类与广义Brownian Sheet的极函数类之间的关系，给出了广义BrownianSheet不动点的Hausdorff维数和Kolmogorov下熵指数的一个不等式。

In the POPF model, wind farm is modeled by the probabilistic wind farm model considering the reactive power-slip characteristic, and the inequality constraints include not only the unit output constraints, the ratio constraints, the voltage constraints and the line current constraints but also the reactive compensation capacity constraints in wind farm and the system climbing capacity constraints per minute. By introducing the NCP function, the KKT conditions of POPF system are transformed equivalently. Based on the transformed nonsmooth nonlinear algebraic equations, the FOSMM is used to determine the POPF model expressed by the numerical characteristic of variables. The model includes nonsmooth functions, so it can be solved by a semismooth Newton-type method based on the subdifferential.

概率最优潮流模型中，风电场采用考虑无功功率—滑加热器差特性的风电概率模型，不等式约束中除了机组出力约束、有载调压变压器变比约束、电压约束和支路电流约束，也考虑了风电场无功补偿容量约束、系统的分钟级爬坡能力约束；使用非线性互补函数将概率最优潮流的KKT条件转化为一组包含有不光滑函数的非线性代数方程组，然后基于一次二阶矩法确定了以待求量的数字特征表示的POPF模型，由于该模型包含不光滑函数，因此采用基于次微分的半光滑牛顿型方加热器法求解。

Using variational trend of key characteristics of population, SAGS algorithm constructed flexible fitness function, cross probability function and mutative probability function.

SAGS算法利用种群关键特征的变化趋势，设计了可变的适应值函数、交叉概率函数和变异概率函数。

Based on the above-mentioned parameter, according to the relationship between robust estimation and nonlinear diffusion, the Turkey loss function is introduced as the diffusion function in SAR image decomposition because of its better performance, and the contour of an image is extracted by the Turkey diffusion function with the proposed diffusion parameter mentioned above. The experiment results indicate that the Turkey loss function based diffusion process can strengthen the conspicuous contour. Additionally, a Raita\'s criterion-based method, solving the automatic diffusion threshold, is proposed to automatically set the threshold in diffusion decomposition.

针对基于梯度参数和局域方差系数的扩散在对SAR图像进行分解时存在的缺陷，提出了一种局域方差系数与窗口幅度均值积的扩散系数，该系数能够更有效地实现潜在目标区域和背景区域的区分；在此基础上，根据鲁棒估计与非线性扩散的联系，引入扩散效果更好的Turkey损失函数作为图像分解中的扩散函数，并结合局域方差系数与窗口幅度均值积的扩散参数来提取图像的轮廓，结果表明：基于Turkey损失函数的扩散过程对特征突出的边缘所起的&强化作用&更加明显；此外，针对扩散分解中的阈值确定问题，提出了一种基于拉依达准则的扩散阈值自动求解方法，实现了扩散阈值的自动求解。

In the negative and interrogative forms, of course, this is identical to the non-emphatic forms.

。但是，在否定句或疑问句里，这种带有"do"的方法表达的效果却没有什么强调的意思。

Go down on one's knees;kneel down

屈膝跪下。。。下跪祈祷