收敛

The study shows the BP neural network model of hydro-turbine for s pecial discharge and special torque can be used to describe its nonlinear charac teristic and ensure the continuity of derivative function of special discharge. The convergence for hammer calculating is improved.

水轮机单位流量和单位力矩特性的BP神经网络模型能真实地表达水轮机的非线性特性，且解决了插值法的导数不连续问题，从而改善了水击计算的收敛性，为用于工程计算和科学分析的水轮机特性表达开辟了新的途径。

The study shows the neural network model of hydro turbine can describe its nonlinear characteristic really and ensures the continuity of its derivative function, so improved the convergence for hammer calculating and is a better description of hydro turbine hill chart for science research and engineering design.

利用所述方法建立的水轮机特性神经网络模型能真实地表达水轮机的特性，并且解决了插值法的导数不连续问题，从而确保了水击计算的收敛性，为用于工程计算和科学分析的水轮机特性表达开辟了新的途径。并利用所建立的数学模型，开发了基于MATLAB的可视化软件仿真平台。

Secondly, based on the Mawhin's continuity theorem in coincidence degree theory and the Brousk's theorem, some novel criteron of the existence on periodic solutions are obtained under the more generalized assumptions than before.

主要工作是：第一，应用构造Liapunov函数以及Lasalle不变性原理，给出了解的有界性与收敛性的一些新的充分条件和充分必要条件，这些结果包含或扩展了已有文献关于同样问题的一些重要结果。

A reinforcement learning algorithm based on linear average is proposed, which is used to solve nonconvergent problems of reinforcement learning function approximation in continuous state space.

提出了一种基于最小线性平均的强化学习算法，用于解决连续空间下强化学习函数估计的非收敛性问题。

A reinforcement learning algorithm based on linear average is proposed, which is used to solve non-convergent problems of reinforcement learning function approximation in continuous state space.

提出了一种基于最小线性平均的强化学习算法，用于解决连续空间下强化学习函数估计的非收敛性问题。

This approach was to replace the derivative operator in the continuous time model by an approximation and estimation of the physical parameters of structures by the least-squares method.

讨论了导数近似值的选择方法，并分析了算法的收敛性。

Applying the Euler method to the equation and particularly using the interpolation technique on variable delay, the Euler approximation is obtained. The discrete Euler approximation is extended to the continuous Euler approximation. It is proved that the Euler numerical solutions will converge strongly to the true solution under the local Lipschitz condition.

对方程应用欧拉方法，特别地对变延迟部分运用插值技巧进行数值离散后，将离散的欧拉格式延拓为连续的欧拉格式，从而得到欧拉格式在局部Lipschitz条件下强收敛到解析解。

Firstly, this paper investigated roundly the basic theory of the ambient vibration method, analyzed systematically the main methods of ambient vibration identification, and discussed their advantages and disadvantages. Then we established the finite element model of structural dynamics and test model and analyzed their mutual relation, improved MUSIC method and correlation function method in the harmonic identification, and gave the subspace method and correlation function method based on continuous model, which is multi-harmonic retrieve technique. And then, this paper also puts forward least-squares method to estimate directly structural physical and modal parameters; this approach replaced the derivative operator in the continuous time model with an equivalent approximation, and estimated the physical parameters of structures by the least-squares method. And then, we discussed the approximation methods of derivative and their convergence. Finally, this paper researched the modal parameters of a frame structure by the simulation and test. The results demonstrated that subspace method, correlation function method and the least-squares method are effective and stable respectively.

全面地研究环境振动识别方法的基本理论，系统地分析环境振动识别方法的主要方法，讨论它们的优点和不足；建立结(来源：6666AB44C论文网www.abclunwen.com)构动力学的有限元模型和试验模型，分析有限元模型与试验模型之间的相互关系；改进谐波识别中的MUSIC识别方法和相关函数识别方法，给出大型结构模态识别的连续型的子空间分解识别方法和相关函数识别方法，这个方法是多元谐波识别方法；提出直接识别结构的物理参数和模态参数的最小二乘方法，这个方法利用导数的近似值代替导数值，并运用最小二乘方法识别结构的物理参数，讨论导数近似值的选择方法，分析算法的收敛性；通过仿真计算和模型实验，研究高层框架结构的模态参数识别，验证子空间分解识别方法、相关函数识别方法和最小二乘识别方法的有效性，算法的稳定性；从理论和试验两个方面分析和研究噪声对识别精度的影响，影响识别精度的因素，以及部分模态成分的模态参数不能够被识别的原因。

As total number of outcomes increases, the sample average of discrete outcomes converges to the probabilistic mean and the sample deviation of discrete outcomes converges to the probabilistic standard divination of continuous random variable.

当出现或试验或取样数量足够的多，则离散随机数值之取样平均值及取样标准差也会收敛为机率平均值及标准差。

Therefore, the voltage/reactive power constrain is considered in the reliability evaluation. The sensitivity algorithm is applied in the adjustment of the reactive power and voltage. The contrariety of the computational effort and accuracy is solved well.

IEEE-RTS 24节点算例结果验证了本文所提出的MCMC方法应用在电力系统可靠性评估中的正确性，相比于传统的蒙特卡洛方法，MCMC方法可以显著提高计算指标的收敛速度，节省计算时间，提高计算的稳定性，并显示出在大型电力系统可靠性评估中具有很大的优越性。

It has been put forward that there exists single Ball point and double Ball points on the symmetrical connecting-rod curves of equilateral mechanisms.

从鲍尔点的形成原理出发，分析对称连杆曲线上鲍尔点的产生条件，提出等边机构的对称连杆曲线上有单鲍尔点和双鲍尔点。

The factory affiliated to the Group primarily manufactures multiple-purpose pincers, baking kits, knives, scissors, kitchenware, gardening tools and beauty care kits as well as other hardware tools, the annual production value of which reaches US$ 30 million dollars.

集团所属工厂主要生产多用钳、烤具、刀具、剪刀、厨具、花园工具、美容套等五金产品，年生产总值3000万美元，产品价廉物美、选料上乘、质量保证，深受国内外客户的青睐