最小

Error of regression model with Least Absolute Deviation is less than with least square method. These two methods are associated in this paper.

本文提出的改进的偏最小二乘回归将这两种方法结合起来，在偏最小二乘回归过程中利用最小一乘法建立因变量对提取的自变量成分的多元回归模型。

This paper presents a method of iteration for least absolute deviation regression,which uses weighted least squares methods.

利用加权最小二乘法，提出最小一乘的一种迭代算法，这种方法使最小一乘法在计算上变得简单、直

By using the principle of least absolute deviation method , we will give a new priority method—least logarithm absolute deviation method.

利用最小一乘法原理，在层次分析中提出了一种新的排序方法——对数最小一乘法，并将其转化成线性规划问题求解，证明了对数最小一乘法的一些性质。

When the random error is not subject to normal distribution, the least absolute deviation estimation is superior to the least squares estimation. In addition, the robustness of the least absolute deviation estimation is also better than that of the least squares estimation.

在随机误差不服从正态分布的问题中，最小一乘估计的统计性能优于最小二乘枯计；另外，最小一乘估计的稳健性更强。

Digital simulations show that the specific fuel consumption was decreased by 5%~7%. The benefit of minimum fuel consumption mode at afterburning condition is much more than that of non-afterburning condition. It has great application potential in engineering.

进行了加力状態最小油耗模式的数字仿真实验，仿真结果表明在加力状態应用最小油耗模式，可以使单位耗油率减少5%~7%，其效益远高於非加力状態的最小油耗模式，具有明显的实际应用价值。

Digital simulations show that the specific fuel consumption was decreased by 5%~7%.The benefit of minimum fuel consumption mode at afterburning condition is much more than that of non-afterburning condition.It has great application potential in engineering.

进行了加力状态最小油耗模式的数字仿真实验，仿真结果表明在加力状态应用最小油耗模式，可以使单位耗油率减少5%～7%，其效益远高于非加力状态的最小油耗模式，具有明显的实际应用价值。

In this paper we have point out the error of principle formula quoted by all literatures at home and abroad to measure the refractive angle and the method of V-prism, derive from beginning the right formula of measuring refractive index by the method of minimum deviation angle and the method of V-prism.

本文指出了测量光学材料折射率的最小偏向角法和V棱镜法国内外所有文献的原理公式误差，重新推导了最小偏向角法和V棱镜法测量折射率的正确公式。基于最小偏向角法用精密测角仅可以获得±3×10-6的折射率的测量精度，然而，这些文献沿用经典的测量原理公式却具有-4×10－4数量级的原理公式误差，因此导致错误的测量结果。

Derive recursive least squares method, recursive general least squares method, recursive augmented matrix method, recursive multiple-stage least squares method and recursive maximum likelihood method based on recursive Householder transformation by applying the three real-time solutions of overdetermined linear equation to difference model.

将线性矛盾方程组的三种实时解法应用于差分模型，导出了基于Househol-der变换的递推最小二乘法、递推广义最小二乘法、递推增广矩阵法、递推多级最小二乘法和递推极大似然法。

In this paper, three robust estimators: robust best invariant quadratic unbiased estimator, robust minimum norm quadratic unbiased estimator and robust Helmert estimator have been designed. Finally, this paper states that these three classical estimators are particular cases of the RBIQUE.

导出了稳健最优不变二次无偏估计、稳健最小范数二次无偏估计、稳健Helmert估计，并说明了最优不变二次无偏估计、最小范数二次无偏估计以及Helmert估计等均是稳健最优不变二次无偏估计的特例。

The monotony and extremum theory of function are applied to prove the geometrical characteristic of error hand for 2D linear segment uncertainty ε(subscript σ) model in theory. The result shows that the error hand has the geometrical shape of "larger ends and smaller middle" or "one end larger and the other smaller". The minimum width and its exact position of the error hand are calculated too. It is proved that the minimum width position of the error hand is near the end with the smaller mean square error.

运用函数单调性和极值理论，从理论上证明ε模型的误差带的几何特征；得到线元误差带不仅具有&两端大、中同小&，而且也可能具有&一端大、一端小&的几何形状；给出了误差带的最小带宽及其位置；论证了误差带的最小误差带宽的位置靠近中误差较小的一端。

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。