AGC

这是matlab中的时频分析工具箱，希望大家用得上。(This is a matlab in time-frequency analysis toolbox, I hope everyone useful.)

说明：  详细讲解了MATLAB语言编程技术，有大量的实例。(Detailed account of MATLAB programming language, there are plenty of examples.)

说明：  我找到的最好的MATLAB教程.目录详细，例子充分！(It s the best selflearning book I ve searched.)

案例2 BP神经网络的非线性系统建模-非线性函数拟合(The BP neural network of nonlinear system modeling-nonlinear function fitting)

程序为ICA的调用函数，输入为观察的信号，输出为解混后的信号(Procedures for the ICA call the function, enter the observed signal, the output signal unmixing)

自己整理的粒子群算法的合集，matlab环境开发的~比较实用(Own finishing particle swarm optimization Collection more practical, the matlab environment for the development of ~)

将从地震勘探工程标准格式的segy文件读出的数据重新写成segy格式，方便用专业的软件进行继续处理。(Reading data seismic engineering standards segy file formats out of the rewritten segy format, easy to use specialized software to continue processing.)

说明：  MATLAB做的STATCOM，用于无功补偿和功率因数校正(power factor corrector using statcom)

小波域高斯混合模型与中值滤波的混合图像去噪研究(Mixed image de-noising in wavelet domain Gaussian mixture model and median filtering)

高速铁路路基沉降是一个复杂的系统过程,涉及不同时期、不同工况,选用单一的预测模型不足以反映路基整体的变形规律.Logistic模型及Verhulst模型的曲线均为S形,可以反映全过程的地基沉降与时间的关系.针对路基填筑期和降水期两个时段利用MATLAB软件编制程序计算模型参数,得到Logistic模型和灰色Verhulst模型,进而求得两者的最优加权组合模型.对比结果表明:在路基填筑期由两者的最优加权组合模型预测得到的结果与实测结果较吻合,拟合精度高于单一模型 而在降水期Verhulst模型的预测值更接近实测值,平均相对误差约为1.75％,拟合精度较高.(High-speed railway subgrade settlement process is a complex system involving different times and in different conditions, the choice of a single forecasting model does not adequately reflect the overall deformation of roadbed. Curve Logistic model and Verhulst models are S-shaped, the whole process can be reflected foundation settlement versus time. against roadbed construction period of two hours and precipitation using MATLAB software programming model parameters, the Logistic model and gray Verhulst model, and then get the optimal model of a weighted combination of the two. comparative results indicate : results and experimental results in the roadbed construction period by the weighted combination of the best of both models predicted than the anastomosis fitting accuracy than a single model precipitation predicted value of Verhulst model closer to the actual value of the average relative error approximately 1.75 fitting accuracy.)