最小

In chapter 1, regression models with censored data, LSE for censored regression models and LAD estimation for models with fixed censoring point are introduced. Then, we state our main results for LAD estimator of nonlinear regression models with randomly censored data, including consistency, boundedness in probability and asymptotic normality. The proofs of our main theorems are given in detail. In chapter 2, we state the main difficulty of obtaining numerical solution for censored LAD estimation, and recall the history of studies on this problem including some anterior algorithms.

在第一章，首先介绍删失线性回归模型的基本概念和性质，并对已有的估计方法，如最小二乘估计、固定点删失的最小一乘估计等做了简单的回顾；然后，把固定点删失线性回归模型的最小一乘估计方法推广到了随机删失、非线性模型的情形，并得到了估计量的一些渐近性质，如相合性、渐近正态性等；最后，给出主要结果的详细证明过程。

An econometric estimation technique involving minimizing the sum of the squares of the vertical distance between the observed data points and the regression line.

最小平方法，最小二乘法计量经济学中的一种估算技术，使观察数据各点和回归线之间的垂直距离的平方之和为最小。

The performance of the final model was evaluated according to the Root Mean Square Error of Prediction and the Correlation coefficient in prediction and calibration sets.

研究了利用联合区间偏最小二乘法和遗传偏最小二乘法等筛选特征谱区的方法，通过交互验证法确定偏最小二乘模型的主成分因子数和筛选区间，并以预测均方根误差和相关系数R作为模型的评价指标。

The mean, maximum and minimum value of the area of transverse section and the major axis length of ellipse and the maximum centroid displacement can be used to discriminate width of pulse tracing.

长轴长度周期平均值、最大值、最小值，面积周期平均值、最大值、最小值，以及轴心在平行于皮肤表面的最大位移等参数可以作为初步判别脉宽的参数。

The agglomerations sizes of fine particles in the fluidized bed are smaller under higher SPL. The sound waves have little effect on the agglomerations when it is under 120dB. But the agglomerations sizes will be dramatically reduced to 180μm from 720μm when the SPL is above 130dB. The umf has a minimum value at an appropriate frequency of sound and a good quality of fluidization can exist in a greater ranges of frequency of sound when the SPL is higher.

当频率高于120Hz，流化床中形成聚团尺寸随着频率的增加而增大，最小流化速度也随着频率的增加而增大，当频率低于120Hz，流化床中形成聚团尺寸随着频率的减小而增大，最小流化速度也随着频率的减小而增大；料层越高，声波的衰减越大，超细颗粒的流化效果越差，最小流化速度越大。

Conceptually, the most obvious sample point within a fundamental period is often referred to as the pitch mark. Usually pitch marks are selected as the local maxima or minima of the audio waveform. In the previous example of pitch determination for the tuning fork, we used two pitch marks that are local maxima. On the other hand, in the example of pitch determination for human speech, we used two pitch marks that are local minima instead since they are more obvious than local maxima.

在观察音讯波形时，每一个基本周期的开始点，我们称为「音高基准点」（Pitch Marks，简称 PM），PM 大部分是波形的局部最大点或最小点，例如在上述音叉的范例中，我们抓取的两个 PM 是局部最大点，而在我的声音的范例中，由於 PM 在局部最大点并不明显，因此我们抓取了两个局部最小点的 PM 来计算音高。

Targeting the biomass rice stalk,we use linestone as catalyst and cracking catalyse the tar which was pyrolysised from the target.We build the rice straw biomass gasification tar removal catalytic cracking process model by the least squares support vector machine model and optimize the model parameters through genetic algorithm.We also has gotten the best catalytic cracking temperature and gas residence time, making the highest rate of catalytic cracking tar.We made wood biomass as a target analysis so that we can thermally crack tar,and we can use genetic algorithms to optimize the model parameters that is based on the wood biomass gasification pyrolysis tar removal process model through the least-squares fitting method,then we got the best pyrolysis temperature and equivalence ratio,so that it is possible to make the tar content reach the minianum.Based on the least squares support vector machine model,we had established the model of the biomass sawdust Compression molding process and fit the Optimization Parameters objective function of the molding process .

以生物质稻秆为对象，采用石灰石作为催化剂对稻秆热解焦油进行催化裂解，由最小二乘支持向量机模型建立了生物质稻秆气化焦油催化裂解脱除过程的模型，并用遗传算法对模型参数进行了优化，得到了最佳的催化裂解温度和气相停留时间，使得焦油催化裂解率达到最高；以生物质木屑作为分析对象，对其进行热裂解脱焦，并依据最小二乘曲线拟合方法建立了生物质木屑气化焦油热裂解脱除过程的模型，用遗传算法对模型参数进行了优化，得到了最佳的热裂解温度和当量比，使得焦油含量达到最小。

The generalized criterion for minimum zone in the determination of form error of surface is presented on the basis of linear minimax techniques, and the criteria of such surfaces as for cylindricity and straightness with arbitrary direction are derived from this generalized criterion.

本文讨论形状误差最小区城的判别准则和判别方法。根据线性极差极小化问题解的特征，给出形状误差最小条件的统一判别准则。推导出各种形状误差最小区域判别法(包括圆柱度误差和任意方向直线度误差)，并在几何判别的基础上得出代数判别的格式。

Then a global separable nonlinear least-squares identification method which estimates the time delays and transfer function parameters separably is derived to reduce the possibility of convergence to a local minimum by using the global optimization theory.Furthermore,the GSNLS method is modified to a novel global separable nonl...

为降低收敛于局部最小的可能性，利用全局优化理论，推导了全局可分离的非线性最小二乘辨识方法；为消除强观测噪声所引起的参数估计的偏差，将GSNLS方法调整为一新颖的全局可分离的非线性多新息递推最小二乘辨识方法，仿真实验验证了算法的有效性。

So the ventilation effect is not good and the heat congregates at housetop in the greenhouse.

结果表明：在时均流场中，互插式连栋温室利用肩窗与侧窗通风时，室外顶部风速最大，天沟处风速最小；温室内部流场风速较小，且顶部风速最小，因而温室室内通风效果不好，热量会聚集在屋顶。

The absorption and distribution of chromium were studied in ryeusing nutrient culture technique and pot experiment.

采用不同浓度K2CrO4(0，0.4，0.8和1.2 mmol/L)的Hoagland营养液处理黑麦幼苗，测定铬在黑麦体内的亚细胞分布、铬化学形态及不同部位的积累。

By analyzing theory foundation of mathematical morphology in the digital image processing, researching morphology arithmetic of the binary Image, discussing two basic forms for the least structure element: dilation and erosion.

通过分析数学形态学在图像中的理论基础，研究二值图像的形态分析算法，探讨最小结构元素的两种基本形态：膨胀和腐蚀；分析了数学形态学复杂算法的基本原理，把数学形态学的部分并行处理理念引入到家实际应用中。