种差

Finishing calculation of mean value, standard deviation, skewness, kurtosis of Beta distribution.(2) Fitting parameters of many kinds of typical distribution and using residual deviation to evaluate fitting precision.(3) Using Beta distribution as an agreed indication distribution applied to many kinds of practical photoelectric measurement distributions.(4) Deriving theory formula of Bayes point estimation about Beta distribution parameters and mean value and standard deviation on the condition of mean square error loss function and supposed the prior distribution is uniform distribution.(5) Generating MCMC sample from post distribution by the method of Gibbs sample algorithm. Calculating bayes point estimation from sample on the condition of mean square error loss function. Calculating confidence interval by an approximate method to complete interval estimation.

本文的主要工作有：(1)解决了Beta分布参数a和b的精确计算以及均值、标准差、偏度、峰度的计算问题；(2)拟合出10余种典型分布的Beta分布的两个参数，并且采用剩余标准差评价该Beta分布的拟合精度；(3)对多种典型的光学与光电测量系统的测量分布进行了Beta分布统示表示；(4)在假设先验分布为均匀分布前提下，得到参数a和b以及均值μ和标准差σ在均方误差损失函数下的贝叶斯点估计理论计算公式；(5)利用直接抽样的Gibbs抽样算法，从后验分布中产生MCMC样本，从样本直接计算均方误差损失函数下的贝叶斯点估计，并使用一种近似方法计算其置信区间，完成区间估计。

The first model is the optimization problems with s linear objective function subject to a system of fuzzy relation equations and a system of fuzzy relation inequalities, which are solved by using path method and 0-1 integer programming with branch and bound method.

第一种为约束为模糊关系方程和模糊关系不等式的具有线性目标函数的优化问题，采用路径法及0-1整数规划中的分枝定界法对其求解；第二种为约束为模糊关系方程和模糊关系不等式的格线性规划问题，采用直接法、路径法及0-1整数规划法对其求解；第三种为模糊关系不等式为模糊的，即约束中有一个可接受的容差，采用容差法及正负理想法对其求解；第四种是一种新的可能性模糊线性规划问题，采用模糊数大小比较法、"切片-粘贴"法及正负理想法，并设计了随机模拟的遗传算法对其求解；第五种为期望值模型，利用期望值将其转化为清晰约束系数的线性规划问题的方法。

The sixteen plants are firstly divided into six types according to evaluation grades and subject function statistical method:the first type is Cajanus cajan and Crotalaria mucronata,the best adaptability ,which can be used pioneer plant or constructive species; the second type is Cynodon dactylon,the better adaptability, which can be used constructive species ;the third type is Eragrostis curvula, Paspalam notatum,which can be used constructive species on rotten rock slope ; the forth type is kummerowia striata,Lotus corniculatus and Vitex negundo.The adaptability is centered,and can be used protection species and auxiliary species;the fifth type is Robinia pseudoacacia,Eremochloa ophiuroides,Vitex negundo,Lespedeza formosa,Leucaena leucocephala,Desmodium sequax, Rhus chinne- nsis,which can't be plant singly and be used accessory species;the sixth is Trifolium repens,which can be used pioneer plant.

根据评价边坡防护植物适应性的十个重要指标，首次运用隶属函数和聚类分析方法，对各供试植物的边坡适应性综合排序并分为六类：第一类为木豆和猪屎豆，边坡适应性最强，适宜作为先锋植物和建群种；第二类为狗牙根，边坡适应性强，适宜与豆科植物混播，作为建群种；第三类为弯叶画眉草和百喜草，边坡适应性较强，适宜与豆科植物混播，作为风化岩边坡的建群种；第四类为鸡眼草、百脉根和马棘，边坡适应性中等，适宜作为保护种与伴生种；第五类为刺槐、假俭草、荆条、美丽胡枝子、银合欢、波叶山蚂蝗和盐肤木，边坡适应性较差，不宜单独在边坡上种植，最好与速生型豆科草本混播，作为边坡防护植物群落的辅助植物；第六类为白三叶，边坡适应性最差，但因其出苗率高、生长速度快，作为先锋植物，起到前期覆盖坡面的作用。

The result indicated that, DTA and DSC apparatus constants of CDR-1 dynamic differential thermal analyzer increase along with temperature increment, but influence of temperature on DTA apparatus constant is more remarkable.

采用6种标准物质测定CDR-1型差动热分析仪的差热分析法和差示扫描量热法仪器常数，结果表明：CDR-1型差动热分析仪的差热分析法和差示扫描量热法仪器常数均随温度的升高而增大，而差热分析法仪器常数受温度的影响更为显著。

In this paper, based on an optimum difference operator for spatial differential given by Holberg, an optimum difference method of modeling crosswell elastic wavefield is put forward, which adopts the staggered-grid technique and calculates the wavefield at different grid points. When the half length of optimum difference operator is taken as four to six points, the required number of grid points per wavelength equals to three, crosshole elastic wavefield numerical modeling can attain satisfied accuracy.

本文以Holberg提出的一种空间微分的最佳差分算子为基础，采用交错网格技术，提出了一种模拟井间弹性波场的最佳差分法，该方法能在大空间网格点上得到精确的波场值，当最佳差分算子的半长度取4～6个点，每个波长只采3个样点时，井间弹性波场的模拟就可以得到满意的精度，其计算速度比常规有限差分法至少快一个数量级。

In order to reduce calculation error, the frequency distribution of average values is used to compute the mixed distribution's digital features of each component distribution, thereinto, the number of the component distribution is determined by AIC, choose the number that meets the minimum value of AIC as the component number of mixed distribution, and the other parameters are estimated by EM algorithm; Secondly, because each component distribution is corresponding to a kind of major gene genotype, according to the values of the average and variance of the each component distribution, we can use the limit error of the normal distribution to plot each individual into the correspondent component distribution, namely into correspondent major gene genotype. Then we regard each major gene genotype as a treatment level of one-way analysis of variances, and the one-way multivariate analysis of variance is carried out to calculate the covariance matrix of major gene effect, covariance matrix of polygene effect, covariance matrix of environment effect and so on; At last, combining the weights of the each component distribution of mixed distribution, we can calculate the variance of major gene effect, the variance of polygene effect, environmental variance and the genetic gain of the quantitative trait.

为减小计算误差，本研究采用均值的频数分布来计算各成分分布的数字特征，其中成分分布个数根据AIC准则，选择使AIC值达到最小的成分分布个数作为混合分布的成分分布数，分布中其它参数的确定利用EM算法来估计；其次，每个成分分布对应一种主基因基因型，根据各个成分分布的均值和方差，利用正态分布的极限误差将每个个体划分到相应的成分分布中，即相应的主基因基因型中，将每种主基因型作为单因素方差分析的一个处理水平，对其进行单因素的多元方差分析，分别计算主基因效应协方差阵、多基因效应协方差阵、环境协方差阵等参数；最后结合混合分布中各成分分布的权重即各主基因基因型的分离比例，计算主基因效应方差，多基因效应方差和环境方差，以及遗传力等参数，进而计算该数量性状的遗传进展。

This project was established based on the first cooperative development of optical in situ real time detecting technique probing the atomic scale layer-by-layer epitaxy growth of oxide film. Over the last three years, we have given full play to the advantages and characteristics of both sides and have finished the assumptions and plans of this subject satisfactorily. 1. In Institute of Physics, we set up advanced oblique-incidence reflectivity difference equipment and wrote one data acquisition procedure that made the simultaneous detection of optical signal and RHEED signal into reality. 2. The first observation of sustained oscillations over hundreds of monolayers in both real and imaginary signals during the epitaxy growth of SrNb0.1Ti0.9O3 on SrTiO3 substrate in LMBE testifies that oblique-incidence is an excellent method to detect and monitor film epitaxy growth real time. 3. For the first time, we verify that oblique-incidence reflectivity difference method can be used to monitor the layer-by-layer growth mode during continuous growth through the correspondence between optical signals to atomic or molecular step edge density on the growth surface. 4. Phenomenological analysis shows that optical signals comprise three parts, the first is proportional to the average thickness of the film and depends on the bulk phase dielectric response; the second is proportional to the coverage of terraces and depends on the dielectric response of atoms or unit cells in the terrace; the third is proportional to the coverage of step edges and depends on the effective dielectric response of atoms or unit cells at step edges. This makes oblique-incident reflectivity difference technique a quantitative macroscopic method to monitor film growth.

中文摘要：本课题是在合作首次发展了氧化物薄膜原子尺度层状外延生长光学原位实时探测方法的基础上立项的，三年来，我们充分发挥双方的优势和特点，圆满完成了课题的设想和计划。1、在物理所建立了一套先进的光反射差法装置，编写了计算机系统的数据采集程序，实现了两路光学信号和RHEED数据的同步采集和显示。2、用激光分子束外延在SrTiO3基底上生长SrNb0.1Ti0.9O3薄膜，首次观测到连续外延几百个原胞层，周期振荡的光反射差实部和虚部信号，证明了光反射差法是一种能原位实时探测与监控薄膜层状外延生长的好方法。3、首次验证了通过对生长表面原子或分子台阶密度的响应，光反射差法可用于原位实时监测在连续生长条件下薄膜的层式生长模式。4、通过唯象理论的研究，证明光反射差信号由三项组成，第一项只与薄膜的平均厚度和宏观光学电介质常数有关；第二项与分子台阶面覆盖度和台阶面上分子层的光学电介质常数有关；第三项与台阶边缘的覆盖度成正比，并和在台阶边缘的分子的有效光学介电常数有关。

Assuming all samples coming from populations with the same standard deviation σ, the accuracy of estimating σ from observations of different parts influences control limits of target control chart. The dissertation points out shortcoming of traditional standard deviation estimates and proposes a unbiased estimate of σ which results in that the variation of the estimate is minimums and the control limits is set more reliably.

假设不同工件的样本标准方差均来自于同一分布，则以不同工件的样本标准方差对母体标准方差的估计的精确性直接影响了目标控制图控制界限计算的可靠性，本文分析了目标控制图传统的两种母体方差估计方法的合理性或稳健性较差的特点，给出了一种对σ无偏估计的优良方法，从而使这一方差估计量〓的散差最小，则依此最小方差的无偏估计量所得到的控制界限更为精确，更为可靠。

A design of temperature sensor with differential output is given, and its sensitivity as the function of temperature is analyzed theoretically and simulated. The theoretical model of response characteristic of viscous damping microbeam system under periodically driven force is built.

在此基础上，进一步从理论上分析了残存应力对微桥固有频率的影响，推导了微桥固有频率同温度的函数关系，提出了一种差动式谐振温度传感器的设计构想，并对其灵敏度随温度的变化进行了理论分析和模拟计算。

A transmission project combining differential gear with axis-fixed gear train s is put forward.

介绍和分析常用齿轮增速箱传动方案，提出一种差动轮系与定轴传动相结合的传动方案，可实现多重变速，扩展系统的测试范围。

The reasons of iron ions content overproof in grade Ⅱ desalting water system,such as variation water quality,contamination of regenerant , operation adjustment of pretreatment system and switching operation of bed were discussed.

对二级脱盐水系统中铁离子含量超标的原因，如来水水质发生波动、再生剂受到污染、预处理系统操作调整、床体运行切换等进行了论述。

You were hired to drum up new business, so go and do it.

公司雇你招徕新业务，你就做你的事好了。