支持向量数据描述故障诊断
所属分类：其他
开发工具：Others
文件大小：355KB
下载次数：2
上传日期：2020-12-16 20:39:47
上 传 者：
2660788648
说明： 支持向量数据描述故障诊断 十分详细的注释
(Fault diagnosis based on support vector data description)
文件列表:
iqiukp-FaultDetection-SVDD-c97d160 (0, 2019-06-02)
iqiukp-FaultDetection-SVDD-c97d160\data (0, 2019-06-02)
iqiukp-FaultDetection-SVDD-c97d160\data\SVDD.png (39534, 2019-06-02)
iqiukp-FaultDetection-SVDD-c97d160\data\X.mat (27972, 2019-06-02)
iqiukp-FaultDetection-SVDD-c97d160\data\Y.mat (6012, 2019-06-02)
iqiukp-FaultDetection-SVDD-c97d160\demo.m (681, 2019-06-02)
iqiukp-FaultDetection-SVDD-c97d160\func (0, 2019-06-02)
iqiukp-FaultDetection-SVDD-c97d160\func\computeKM.m (3378, 2019-06-02)
iqiukp-FaultDetection-SVDD-c97d160\func\normalize.m (1039, 2019-06-02)
iqiukp-FaultDetection-SVDD-c97d160\func\plotResult.m (1137, 2019-06-02)
iqiukp-FaultDetection-SVDD-c97d160\func\svdd_opt.m (799, 2019-06-02)
iqiukp-FaultDetection-SVDD-c97d160\func\svdd_smo.m (1371, 2019-06-02)
iqiukp-FaultDetection-SVDD-c97d160\func\svdd_test.m (667, 2019-06-02)
iqiukp-FaultDetection-SVDD-c97d160\func\svdd_train.m (1547, 2019-06-02)
iqiukp-FaultDetection-SVDD-c97d160\ref (0, 2019-06-02)
iqiukp-FaultDetection-SVDD-c97d160\ref\Tax-Duin2004_Article_SupportVectorDataDescription.pdf (318947, 2019-06-02)
www.downma.com.txt (151, 2017-12-14)
## Fault detection based on support vector data description (SVDD)
-------------------------------------------------------------------
Updated on 2 Jun 2019
1. Added the SMO algorithm to solve the Lagrange dual problem of SVDD
-------------------------------------------------------------------
Updated on 28 May 2019
1. Added descriptions for the calculation of the radius
2. Added descriptions for the calculation of the distance
from the testing sample to the center of the sphere
-------------------------------------------------------------------
-------------------------------------------------------------------
Updated on 24 May 2019
1. Fixed some function descriptions
2. Added support for some common kernel functions
-------------------------------------------------------------------
## demo
```
% DESCRIPTION
% Fault detection based on Support Vector Data Description
%
%
% ---------------------------------------------------------------------%
clc
close all
addpath(genpath(pwd))
% Load data (X: training data Y: testing data)
load ('.\data\X.mat')
load ('.\data\Y.mat')
% Normalization (in general, this step is important for fault detection)
[X_s,Y_s] = normalize(X,Y);
% Set parameters
C = 0.2; % trade-off parameter
s = 6; % kernel width
ker = struct('type','gauss','width',s);
% Train SVDD hypersphere
model = svdd_train(X_s,C,ker);
% Test a new sample Y (vector of matrix)
d = svdd_test(model,Y_s);
% Plot the results
plotResult(model.R,d)
```
近期下载者：
相关文件：
收藏者：