File Exchange

image thumbnail

Kernel Principal Component Analysis (KPCA)

version 2.0 (3.47 MB) by Kepeng Qiu
KPCA for dimensionality reduction, fault detection, and fault diagnosis.

62 Downloads

Updated 16 Nov 2019

GitHub view license on GitHub

KPCA for dimensionality reduction, fault detection, and fault diagnosis.

1. For a brief introduction to this code, please read 'contents.m' file.
2. Fault diagnosis module only supports gaussian kernel function at this version.
3. The fault diagnosis module calls the precompiled file '.mexw64' to accelerate the running speed.
4. Class is defined using 'Classdef...End', so this code can only be applied to MATLAB after the R2008a release.

Please go to https://github.com/iqiukp/Kernel-Principal-Component-Analysis-KPCA for details.

Comments and Ratings (3)

enjian cai

Shanfei Su

Thanks for your great algorithm of KPCA(with Fault Diagnosis). It's owesome to solve the nonlinear problem.
When I rethink the algorithm, I think of that the algorithm belongs to Static KPCA,what if develop a dynamic KPCA method? It seems that there will be a more accurate diagnosis. So I try it.
But I failed to finish it (maybe because I do a wrong thing while making the augmented matrix ).
Sincerely, can you make a further step to realize the algorithm of dynamic KPCA?
Thanks again for your great work!

Here are some references:
1. Mingxing Jia ⁎, Fei Chu.On-line batch process monitoring using batch dynamic kernel principal component analysis(J).
2. Ines Jaffel a, OkbaTaouali. Moving window KPCA with reduced complexity for nonlinear dynamic process monitorin(J).
3. YUAN Zhe,SHI Huaitao. Fault Diagnosis Approach Based on Step Dynamic KPCA(J)

P.S. I also followed you at Github.

Updates

2.0

1. Used OOP to rewrite some modules.
2. Added support for multiple kernel functions.
3. Fixed some bugs.
4. Accelerated the running speed of the fault diagnosis module.

1.2

1. Fixed some errors
2. Added Dynamic KPCA(DKPCA)

1.1.1

1. Fixed some errors
2. Added fault diagnosis

1.1

1. Fixed some errors
2. Added fault diagnosis

MATLAB Release Compatibility
Created with R2019b
Compatible with R2018a to any release
Platform Compatibility
Windows macOS Linux