SVM

用abaqus单元开发子程序uel生成弹簧单元。(User subroutine VUEL defining a three-dimensional spring element with degrees of freedom ordered in a nonstandard way)

复数的四则运算完整版，包括加减乘除，阶乘，开方等。。。(The plural arithmetic full version,, including addition, subtraction, multiplication, and division factorial, prescribing, etc...)

简述算法在实际当中的应用以及算法的改进和延伸(Improvement and extension of applications and algorithms of the outlined algorithm in practice)

这是一本关于大型有限元模拟软件ABAQUS的英语书籍，从入门介绍到应用实例，还可以学习英文。如果你需要你一定会喜欢的。(English contents: 1. Introduction 2.Bar Element 3.Beam Element 4.Rigid Jointed Frames 5.Stress and Strain Analysis 6.Weighted Residual Methods 7.Finite Element Approximation 8.Numerical Integration 9.Plane Problems 10.Axisymmetric Problems 11.Thin and Thick Plates )

Finding special triples of the form (n, n + 1, n + 2) Write a program that finds all triples of consecutive positive three-digit integers each of which is the sum of two squares, that is, all triples of the form (n, n + 1, n + 2) such that: • n, n+1 and n+2 are integers at least equal to 100 and at most equal to 999 • eachofn,n+1andn+2isoftheforma2+b2.(Finding special triples of the form (n, n+ 1, n+ 2) Write a program that finds all triples of consecutive positive three-digit integers each of which is the sum of two squares, that is, all triples of the form (n, n+ 1, n+ 2) such that: • n, n+1 and n+2 are integers at least equal to 100 and at most equal to 999 • eachofn,n+1andn+2isoftheforma2+b2.)

利用AR模型进行时间序列预测的程序源代码，使用最小二乘估计法进行参数估计。拟合效果非常好。(use AR model for time series prediction of the source code, the use of least squares estimation method to estimate parameters. Fitting very good results.)

针对多径效应的影响，提出了一种基于矩阵束的MIMO 雷达低仰角快速估计方法。该方法同时考虑了发射多径信号和接收多径信号，采用单样本数信号矢量构造了一个前后向矩阵束，并利用两个酉矩阵对该矩阵束进行降维处理，最后采用广义特征值分解的总体最小二乘法来估计目标角度。算法不需要估计数据协方差矩阵，可在低 信噪比和单样本数情况下，有效地克服多径效应，实现同时多目标低仰角估计，相比最大似然算法，避免了谱峰搜索，计算量小。仿真结果验证了该算法的有效性。(To overcome the multipath effect, a fast algorithm for low elevation estimation for MIMO radar via matrix pencil is proposed. The signals both in transmit multipath and receive multipath are considered. Firstly, a forward-backward matrix pencil is formed based on the single-sampled vector. Secondly, the reduced-dimensional processing is applied to the matrix pencil via two unitary matrices. Finally, the generalized eigen-decomposition is employed to estimate the low elevation in multipath environments directly. Without the covariance matrix estimation, the proposed method can overcome multipath effect effectively in case of low SNR and estimate multi-target simultaneously. Compared with Maximum Likelihood (ML) method, it avoids spectrum peak searching and reduces the computational burden. Numerical results verify the effectiveness of this method)

牛顿迭代法求解非线性代数方程组的主要思想是将非线性函数线性化(Newton iteration method for solving nonlinear algebraic equations nonlinear function of the main idea is to linear)

说明：  Neville插值算法，xa为自变量数组，xy为因变量数组；x为待插值的自变量(Neville interpolation algorithm, xa array as variables, xy array as the dependent variable x for the independent variables to be interpolated)

利用QS4算法解决N皇后问题，解决300万的皇后所需时间在5s左右，特别快(Use QS4 algorithm to solve the N-queens problem, solve 3000000 Queen the time required for about 5s, particularly fast)