算法

This algorithm not only avoids the local optima and also robust to initialization.

利用改进遗传算法强大的全局寻优能力，这种算法较好地克服了FCM算法对初始化敏感、容易陷入局部最优的缺陷。

In order to overcome the shortagings of current algorithms, a new learning algorithm called as"Recurrent Generalized Inverse Peneaty" algorithm is derived via the optimality principle.

主要结果有：（1）根据最优化原理提出了新的训练GSLN网络的学习算法，"递推广义逆罚函数"算法，以便克服现有算法的缺陷；（2）。

In this paper,the scheme of autonomous orbit determination of satellite is described,the extended Kalman filter algorithm based on improved numerical method is put forward.

高精度的自主定轨对探测器实现深空探测任务有着重要的意义，结合利用星光折射角的测量确定卫星轨道的自主导航方案，提出了基于数值算法改进的扩展卡尔曼滤波算法，提高了运算精度，通过仿真验证了该算法的优越性。

Considering the effect of interval on the original estimation accuracy and noise factors, the optimal interval, which minimizes estimation errors, is derived. It is related with MT's velocity. In order to reach the optimal interval, joint velocity estimation and interval calculation are applied. Finally, the feasibility of proposed algorithms is discussed.

通过对AWGN信道下的COV速度估计精确建模，数值分析了协方差计算间隔对COV算法本身精度及噪声因子的影响；根据分析的结果，以归一化平均误差最小为目标对估计算法进行了改进，获得了基于COV的精确速度估计；最后还对算法的可行性进行了讨论。

Furthermore, the paper presents a line oriented locating algorithm for fast locating triangles, which has the advantage of unique path and steady operation compared with other locating algorithms.

同时提出了快速定位三角形的直线方向定位算法，该算法与一般方向定位算法相比，具有搜索路径唯一，运行稳健的特点。

Thesis totally five, chapter 1 just introduces the background that the calculate way dynamic state imitates to play to show the system, chapter 2 explained the FLASH of meaning, however the most is chapter 4 of Sha-ho, introduce I create the process that the dynamic state emulation plays to show the system,such as the realization of the row preface calculate way, and check to seek the calculate way of realization, home page establish, the creation, orotund ducting and the constitution of the button

论文一共五章，第一章只是介绍算法动态模拟演示系统的背景，第二章说明了一下FLASH的意义，不过最主要的是第三和第四章，介绍了我制作动态模拟演示系统的过程如排序算法的实现、查找算法的实现、首页的创建、按钮的制作、声音的导入和设置。

In this paper, we discuss the computational stability of two kinds of algorithms for solving Least-Squares problem, the normal equation method and the orthogonal decomposition methdd, and show that we can get better performance in Jater case in this aspect.

其一是基于正则方程的递推求解方法，这是目前各种自适应算法的基础；另一是正交分解算法。后者的计算稳定性要比前者高。研究了一种基本的正交分解自适应算法，并指出，如果初始条件恰当，它将得到和批处理同样的结果。

This thesis discussed the recursive least square algorithm , M-P pseudo-inverse algorithm and recursive orthogonal least square algorithm for the RBPNNs.

文中讨论了递推最小二乘算法、M-P伪逆算法和递推正交最小二乘算法用于径向基概率神经网络的训练问题。

The experiments in the ORL face database showed that the proposed algorithm is characterized by a higher recognition rate than the traditional algorithms for the global orthogonal and the uncorrelated discriminant vectors sets, and the highest recognition rate may reach 99%.

在ORL人脸库中的测试结果表明，与传统的全局正交鉴别矢量集算法及传统的无相关鉴别矢量集算法相比，基于核映射的无相关鉴别矢量集算法有更高的识别率，最高识别率可达到99%。

Through the study of the theories about the C tool radius compensation, the dissertation put forward a new algorithm of C tool radius compensation for the effective reduction of overcut while NC-machining. C++ programming language is adopted to realize the algorithm, and examples of simulation are given to demonstrate the achievements of the algorithm.

根据刀具半径补偿原理，实现了一种新型的刀具半径补偿算法，该算法能有效剔除冗余刀位点，并且用C++实现了该算法，给出了相应的仿真实例。

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。