在算法上

It includes as follows:(1) That the right acceptance probability to decide the initial value temperature should be 0.6 are presented by experiment. For the first time, crossover and mutation operator are used in populations to calculate the initial value of temperature. At least the initial value of temperature is 〓.(2) Replacing random change operator with crossover and mutation operators.

2使用遗传算法中交配与突变代替模拟退火算法中随机扰动操作，对两种操作方法进行定性分析，得到突变和交配操作与随机扰动方法在本质上的一致性；同时首次提出在不同温度条件下，使用不同的比例因子，控制参与交配与突变的基因位数，实现了二次使用模拟退火算法的过程。

Also the problems of stochastic controllability, observability and stability of the filter are studied, a simple observability criterion for the time-invariant systems is set up, error band of the filtering is given theoretically, and the filtering stability results are proved under more general conditions; besides, the filtering error caused by uncorrect mathematical models is analysed.

本文还研究了随机能控能观性及滤波的稳定性问题，建立了结构简单的平稳定常系统的能观性判据，在理论上给出了滤波算法的误差范围，在更一般的条件下证明了最优滤波算法的稳定性；并从理论上分析了模型误差对滤波产生的影响。

After introducing the primary approximation approaches for curve and surface offset, we give an optimal approximation approach for offsetting curve based on shifting control points, which relies on approximating the normal by Legendre polynomial approximation.

在全面介绍了等距曲线曲面逼近的主要方法以后，作者首先利用Legendre多项式得到基曲线法矢曲线的最佳平方逼近，由此给出了基于控制顶点偏移的等距曲线最优逼近的算法，它在算法效果和效率上都远优于其他同类算法。

According to theoretical analysis, it is convergent and has good convergence if the coefficient matrix is a nonsingular matrix. Moreover, examples have been implemented on HP rx2600 cluster, and the data verify that the numerical results of this algorithm coincide with its theoretics.

在理论上，算法只要系数矩阵A是非奇异矩阵就能够收敛，而且有较好的并行性，在HPrx2600集群上进行数值试验，验证了算法的计算结果与理论相一致。

In order to improve the efficiency, we present two new encoding/decoding schemes, whose performance is better than that of the classical algorithm when they are applied to quantize high-frequency WT coefficients.

为了提高熵编码的效率，本文在经典算法的基础上提出了两种新的编码方案，并应用于小波变换编码中，在性能上优于原算法。

The traditional genetic algorithms have a set of relatively perfect algorithm systems and have been applied in many optimization problems successfully, but there still exist some drawbacks such as lack of local search ability, premature convergence and random walking etc in the algorithms themselves, which lead to the weak capability of convergence.

传统的遗传算法虽然在理论上已经形成了一套较为完善的算法体系并在许多优化问题中都有成功的应用，但其本身还存在一些不足，例如局部搜索能力差、存在未成熟收敛和随机漫游等现象，从而导致算法的收敛性能差。

This method took the correlation of every color component into account reasonably, and converted the computation of a vector space to scalar quantity in a natural way.

这种边缘检测方法，在一定程度上合理地考虑了个各颜色分量的相关性，将向量空间的计算以自然的方式转换成了标量的计算，在思想上是一种不同于其他算法新的算法。

The simulative experiment is done by Matlab, and it is shown that BISK has higher performance and better image compression quality than OR-SPECK.

在Matlab中进行的仿真实验表明，在对任意形状对象进行编码时，BISK编码算法在算法性能和图像压缩质量上较OB-SPECK编码算法更优。

The upper bound of error probability is used as the criteria of clustering and the evaluation of the clustering result. On the basis of initiatory clustering based on distance functions, the algorithms named"pick-over"and"fill-up"are proposed to minimize the upper bound of error probability.

我们将聚类准则即模式相似性测度问题和聚类结果的评价指标统一为总的最小错误概率上界，在算法实现上，首先先用基于距离函数的聚类方法得到初分类，然后采用基于最小错误概率的&抽取&和&回填&算法得到使总的错误概率上界最小的聚类结果的分析方法。

Prediction the protein subnuclear locations; c we have for the first time extended the prediction scope to cover 22 subcellular locations, which greatly improves the practical value of the computational models.

本文首次在国际上提出并探讨了a 蛋白在细胞中多个位置出现的预测模型；b 蛋白在细胞核中出现的位置的预测模型，即&亚亚细胞位置预测模型&，获得国际学术界的认可；c 本文首次将亚细胞定位的预测研究推广到覆盖22个亚细胞位置，极大地提高了预测模型的实用价值，并提出了融合蛋白序列高层基因本体特征及序列自身氨基酸特征的蛋白亚细胞位置预测方法，提出了面向不同物种的亚细胞定位的预测新思路；结果表明新算法方法在严格的数据集上获得了比传统算法方法高出35%以上的预测精度，所开发的工具被广泛应用于生物实验中。

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。