An improved moving window kernel principal component analysis algorithm

SUN Jianping; LI Xue; LIU Xuan; LIU Chen

English

您当前的位置：

首页 >

文章列表页 >

An improved moving window kernel principal component analysis algorithm

Thermal Power Generation Issue 12, (2017)

更新时间：2025-09-28

- An improved moving window kernel principal component analysis algorithm
- 作者机构：
  
  1. 华北电力大学控制与计算机工程学院,河北,保定,071003
  2. 中国移动邯郸分公司,河北,邯郸,056000
  3. CNNC Xinneng Nuclear Engineering Co., Ltd.,, Ch,Taiyuan,030015
  4. 中核新能核工业工程有限责任公司,山西,太原,030015
- 作者简介：
- 基金信息：
- DOI：
  CLC： TM614
- Published Online：04 January 2018，
  
  Published：2017
- 稿件说明：
移动端阅览
SUN Jianping, LI Xue, LIU Xuan, et al. An improved moving window kernel principal component analysis algorithm[J]. Thermal Power Generation, 2017, (12).
DOI：

SUN Jianping, LI Xue, LIU Xuan, et al. An improved moving window kernel principal component analysis algorithm[J]. Thermal Power Generation, 2017, (12). DOI：

摘要

传统滑动窗口主元分析算法在处理工程参数缓慢劣化问题时存在过适应现象，且容易产生误报、漏报的问题。为此，本文以滑动窗口核主元分析（KPCA）算法为基础提出一种基于数据块的改进滑动窗口核主元分析算法。该算法采用MATLAB软件进行数据标准化预处理，利用迭代法剔除采样数据异常值；在检测过程中以数据块为单位进行模型更新，通过调节数据块的大小，使故障诊断更加快速、准确。将该算法和传统核主元算法分别应用于风力发电机组实验数据分析，验证了本文所提出的改进算法能够更好地解决参数缓慢劣化问题。

Abstract

关键词

Keywords

references

Views

500

下载量

CSCD

Alert me when the article has been cited

提交

Publicity Resources

A fault diagnosis method for coal mills based on Bayesian network under imbalanced data conditions

Review of fault monitoring and diagnosis for induced draft fans in power stations

Function design and application exploration of intelligent DCS for coal-fired power plants

A rotor fault diagnosis method based on EEMD cloud model and SVM

A survey of condition monitoring and fault diagnosis for wind power generators

Related Author

胡绍宇

郝欣

刘乐源

邵毅

张涛

邓楷文

李明昊

何新

Related Institution

Key Lab of Condition Monitoring and Control for Power Plant Equipment, North China Electric Power University

Xi'an Thermal Power Research Institute Co.,Ltd.

Huadian Electric Power Research Institute

Huadian Power International Corporation Limited

School of Energy, Power and Mechanical Engineering, North China Electric Power University

AI问答

⁰