space_error_model

杜芬振子相轨图在利用混沌处理机械故障微弱信号应用Duffing相轨(Duffing oscillator with mechanical failure orbit map at a weak signal processing phase Duffing orbit)

Generates the coordinates necessary for plotting an automatically sized 2D spring

对连续的单一频率离散序列 ，即k=8。 按采样频率 采样，截取长度N分别选N =20和N =16，观察其DFT结果的幅度谱。 (Discrete to continuous sequence of single frequency, ie k = 8. According to the sampling frequency sampling, truncation length N = 20 N, respectively, and selected N = 16, observed the results of its DFT magnitude spectrum.)

Either a string naming a function that calculates the gradient, or the null matrix

nmf的两个简单例程，搜集于网络，为matlab代码。(nmf two simple routine to collect network matlab code.)

分类与回归。Matlab的工具箱是围绕一个快速LS-SVM训练和模拟算法。该相应的功能调用可以用于分类以及为函数估计。该功能plotlssvm显示模型的模拟结果，在训练的区域点。(The Matlab toolbox is built around a fast LS-SVM training and simulation algorithm. The corresponding function calls can be used for classification as well as for function estimation. The function plotlssvm displays the simulation results of the model in the region of the training points.)

时间序列数据分析中的梅林变换工具，数据包传送源码程序，wolf 方法计算李雅普诺夫指数。( Time series data analysis Mellin transform tool, Data packet transfer source program, wolf calculated Lyapunov exponent.)

Linar feed back register matlab code... Another matlab routine!!!

说明：  MGM(1,1)模型预测及精度检验，可直接使用(MGM (1,1) model prediction and accuracy test can be used directly)

第一章 概述 辨识（Identification）、状态估计和控制理论在现代控制理论中是密不可分的，它们互相渗透。辨识和状态估计离不开控制理论的支持，控制理论的应用几乎不能没有辨识和估计技术。由此可见，辨识在科研与实际应用领域都有很高的实用价值。 所谓辨识，就是从含有噪声的输入、输出数据中提取被研究对象的数据模型。此数据模型只是过程的输入输出特性在某种意义下的近似，而近似的准确度一般取决于采样数据的精度以及辨识方法的合理性。 辨识的目的是根据过程所提供测量的数据等信息，在某种准则意义下，估计出模型的未知参数，其基本原理如图1.1所示： (Identification (Identification), state estimation and control theory in modern control theory are inextricably linked, they penetrate each other. Identification and control of state estimation theory is inseparable from the support of the application of control theory can hardly no identification and estimation techniques. This shows that the identification of the research and practical application areas have high practical value. The so-called identification, is to contain noise from the input and output data from the study was the data model. This data model is the process of input and output characteristics in a certain sense of approximation, while similar to the general accuracy of the data depends on the accuracy of sampling and identification method is reasonable. Identification wit)