参数假设

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。

A new non-linear quadratic programming Bayesian prestack three-terms inversion method is developed; Firstly, this method is based on Bayesian parameter estimation theory, Gaussian distribution is used for likelihood function and modified Cauchy distribution is used for prior distribution; Secondly, covariance matrix is used to describe the degree of correlation between the parameters; rock physics relations are used to constrain the inversion results; at last, this method is transformed into non-linear quadratic programming problem, and inversion results are acquired under several constraints.

本文提出了一种基于非线性二次规划的叠前三参数反演方法。首先基于贝叶斯参数估计理论，假设似然函数服从高斯分布，并使待反演的参数服从于改进的Cauchy分布，从而提高了反演结果的分辨率；其次用协方差矩阵来描述参数间的相关程度，进一步提高了反演结果的稳定性；最后将问题转化为一个非线性二次规划的求解问题，并在多种约束下得到问题的解。

The thesis can be divided into five parts as follows: First, the structural characteristics of finned-pipe evaporator are analyzed. After selecting suitable microelement controller, the heat-transfer and mass-transfer processes are analyzed for every microelement under the conditions of dryness, wetness and frostiness. Based on previous equations, some parameters of frostiness are confirmed and the frost-growing model is set up under frost condition. Some hypotheses are postulated and with the help of the equation of mass-conservation, energy-conservation and momentum-conservation, the evaporation model which fits in the dynamic simulation is built, which set a solid foundation for system simulation. Second, the starting and stopping behaviors under disturbed condition are analyzed and calculated by using the dynamic concentrative parameter model, which gives some advice to better prescribe refrigeration system and set theoretic foundation for carrying out automatic control of refrigeration system. Third, the normal running process is analyzed and calculated by means of rational matching theory, which gives some advice on how to better understand the parameter change under steady state and the affection of inlet-parameter on evaporator. Fourth, the simulation software with dynamic characteristic is designed, which can be applied to calculate thernio-parameter of cryogen, air humidity and frost thickness under different initial and boundary conditions, and to carry out dynamic simulation under conditions of dryness, wetness and frostiness, at the same time, to achieve detection and simulation at any stage from starting to stopping.

本文的主要内容如下：1对翅片管蒸发器结构特点进行分析，选取适当的微元控制体，就干、湿和霜工况下对每个微元分别进行传热传质分析，基于经验关系式确定霜的有关参数，对于霜工况下的霜生长建立模型，经适当假设，运用质量守恒、能量守恒和动量守恒方程建立适合动态仿真的蒸发器数学模型，为系统仿真奠定基础； 2对蒸发在大扰动下的开、停机过程，运用动态集中参数模型进行分析和计算，为更好地描述制冷系统运行的全过程奠定基础，同时也为制冷系统实现自动控制提供一定的理论基础； 3对蒸发器正常运行过程，运用动态分布参数和参数间定量耦合的观点来分析和计算，为更好地了解稳态工况下各点参数的变化情况及各入口参数对蒸发器动态特性的影响即蒸发器性能对各参数变化的敏感性； 4编写翅片管蒸发器动态特性仿真计算程序，可以计算不同边界条件和初始条件下的制冷剂热力参数、空气温湿度和霜厚度分布场，实现对翅片管蒸发器在干、湿和霜工况下的动态仿真。

According to Lemaitre's damage evolution law and the assumption of this paper that the cyclic loading maximum tensile stress has the most contribution to the cyclic damage and assume that the stress is the maximum stress .The material parameters S which are expressed in exponential form and generalize the variation in the process of cyclic loading is exponential experience equation which provide the expectation of cyclic damage of 63Sn/37Pb material and lead it into the calculation of damaged incremental form and assume that damage keeps constant during the same cycle. The results match the experimental data. It proves that the Endochronic plasticity theorem adopt the calculation of damage environment.

根据Lemaitre损伤演化律及本文假设循环负载下拉应力在最大值附近对循环损伤的贡献最大，文中应力对各循环之损伤而言假设应力为最大拉应力，将材料参数S以指数型式表示，归纳出疲劳损伤在循环负载过程中的变化为指数型经验式，提供63Sn/37Pb材料在循环负载作用时损伤情况的预测，并引入含损伤增量式内涵时间之计算中，假设同一循环的损伤为定值，其计算结果与实验数据吻合，说明内涵时间塑性理论亦适用於损伤环境下的计算。

Hierarchical clustering of multivariate data or distance data; disjoint clustering of large data sets; nonparametric clustering with hypothesis tests for the number of clusters.

聚类多元数据或距离数据；大型数据集不相交的聚类；与集群的若干非参数假设检验集群。

To recover the time function of source, the deconvolution procedures based on least square method were developed, and to obtain the Frechet derivative of this function, we assumed it as a polynomial of tenth order, then the time function can be treated as a parameter to be identified.

至於波源时间之反求，则以最小二乘法为之，并假设其为十次多项式，而由此求出Frechet导数，并将其视为欲辨识之参数之一，由所得之结果得知，辨识出来之参数和事先假设者颇为相符。

In this paper,a revised universal soil erosion e-quation was selected, the effects of errors of this neglection on it were tested by non-parametric hypothetical test method of Kolmogorov-Smirnov for a 22-year period within 21 catchments of the middle Huanghe River region,and the resul...

采用非参数假设检验方法，对年产沙模数估计方程中下垫面因素的时变影响作定性分析，结果表明，对于一般流域，如果没有人类活动或自然灾害破坏导致下垫面条件的突变，下垫面因素对产沙的影响可看作是时不变的。

In view of the peaked and fat-tailed characteristics of financial return data distribution and its effect of clustering fluctuation and especially the "leverage effect" of fluctuation on VaR estimates and some efficiencies when estimating VaR with various assumptions of return data distribution,a semi-parameter approach based on EGARCH-VaR model is developed.

在综合考虑了金融收益数据分布的尖峰厚尾特征及其波动集群性，尤其是其波动的&杠杆效应&对VaR估计的影响以及各种假定收益率分布在计算风险价值时存在不足的基础上，提出了基于EGARCH-VaR的半参数方法，并且与正态分布和t分布假设下的GARCH模型的VaR计量方法进行比较，通过实证分析，并利用后验测试，表明基于EGARCH-VaR的半参数方法对风险价值的测度优于正态分布和t分布假设下GARCH模型的VaR计量方法。

And objective parameters (texturized degree, color, hardness, springiness, chewing, water absorption, yield, etc.) were investigated in the high moisture soy protein extrusion process, and the statistical models of system and objective parameters were built, using the step-by-step regression analysis; A comprehensive evaluation of the products, and the process optimization of high moisture extrusion were obtained,using factor analysis; Secondly, residence time distribution and soy isoflavones losses kinetics were studied in the high moisture extrusion; Finally, chemical bonding process, the micro-structure and protein secondary structure changes were investigated in the texturization of soy protein, then the mechanism assumptions of texturization soy protein by moisture extrusion were proposed.

本文对大豆蛋白高水分挤压组织化技术和机理进行了研究，研究内容包括：首先采用系统分析法，研究了大豆蛋白高水分挤压过程中，挤压机操作参数(螺杆转速、物料水分、喂料速度和机筒温度)对系统参数(系统压力、扭矩、单位机械能等)、目标参数(组织化度、色泽、硬度、弹性、咀嚼性、吸水率、产量等)的影响规律，建立了各因变参数的统计模型；采用因子分析法对产品进行了综合评价，并对高水分挤压组织化工艺进行了优化；其次，研究了大豆蛋白高水分挤压中的停留时间分布以及大豆异黄酮的损失动力学；最后，研究了大豆蛋白组织化过程中化学键、微观结构和蛋白质二级结构的变化规律，提出了大豆蛋白高水分挤压组织化的机理假设。

The method to configure RED parameters based on network properties is also presented. Simulations show that the modified Random Early Detection algorithm really performs better than the original Random Early Detection algorithm and can work for a much wider range of traffic. Finally, whether the assured service mechanisms i. e. an edge router tags every arriving packet of a TCP flow based service profile and core router drops packets differently at congestion using active queue management scheme is possible to provide a throughput consistent with target rate, is investigated under different network scenarios. Different factors including target rate tagged algorithm, RIO algorithm and TCP congestion control mechanism how to affect the throughput of an assured service TCP flow are examined. Based on a fluid model. we then derive a model of an assured service TCP flow, which is the function of round trip delay, packet loss rate and token bucket parameters. Based on this model, we observe that in some cases it is possible to regulate the token bucket parameters to gain a consistent throughput with the target rate, while in other cases there exist ranges of values of the achieved rate for which the token parameters have no influence. With the latter, some modified TCP congestion control mechanisms to assure TCP throughput are proposed.

最后，在区别服务网络结构内，研究了当前提出的一种边缘路由器根据服务合同标记、分类数据包，结合内部路由器采用主动队列缓冲管理机制区别处理数据包的确信服务机制是否可在不同网络情况下使TCP连接获得同目标速率相一致的、公平的吞吐量问题；指出影响TCP连接获得同目标速率相一致吞吐量的主要因素是目标速率大小、标记算法参数和TCP连接的拥塞控制算法；对此，基于流模型假设，首次推导出一个以端到端时延、数据包丢失率、漏桶参数为变量的确信服务TCP连接吞吐量模型；利用这个模型进行分析，得到主要结论是在一些情况下可通过设置合适的漏桶参数得到同目标速率更一致的吞吐量，而在另一些网络情况下不管如何设置漏桶参数都不能使TCP连接获得同目标速率一致的吞叶量；对于后一些情况，还初步讨论了改进TCP拥塞控制算法来改善TCP连接性能的方法。

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。