[1]汪国新,郝勇生,苏志刚.基于KPCA-GG的火力发电设备状态诊断方法[J].东南大学学报(自然科学版),2019,49(3):542-548.[doi:10.3969/j.issn.1001-0505.2019.03.020]
 Wang Guoxin,Hao Yongsheng,Su Zhigang.Condition diagnosis method for thermal power generation equipment based on KPCA-GG[J].Journal of Southeast University (Natural Science Edition),2019,49(3):542-548.[doi:10.3969/j.issn.1001-0505.2019.03.020]
点击复制

基于KPCA-GG的火力发电设备状态诊断方法()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
49
期数:
2019年第3期
页码:
542-548
栏目:
自动化
出版日期:
2019-05-20

文章信息/Info

Title:
Condition diagnosis method for thermal power generation equipment based on KPCA-GG
作者:
汪国新1郝勇生12苏志刚2
1东南大学计算机科学与工程学院, 南京210096; 2东南大学能源与环境学院, 南京210096
Author(s):
Wang Guoxin1 Hao Yongsheng12 Su Zhigang2
1School of Computer Science and Engineering, Southeast University, Nanjing 210096, China
2School of Energy and Environment, Southeast University, Nanjing 210096, China
关键词:
核主成分分析 多元时间序列分割 Gath-Geva模糊聚类算法 火力发电设备
Keywords:
kernel principal component analysis(KPCA) multivariate time series segmentation Gath-Geva fuzzy clustering thermal power generation equipment
分类号:
TP274
DOI:
10.3969/j.issn.1001-0505.2019.03.020
摘要:
为了解决具有非线性特征的设备状态诊断问题,提出一种基于核主成分分析和Gath-Geva模糊聚类相结合的多元时序分割算法.根据Gath-Geva模糊聚类算法得到聚类结果,利用核主成分分析算法提取非线性特征,从而构造KPCA分析模型.将聚类类簇在该模型空间中的距离作为类簇相似性分析及合并的标准,以提升方法的分割效果.实验结果表明,基于KPCA的Gath-Geva模糊聚类算法能识别数据的非线性信息,更准确地分析数据特征,其分割效果优于基于主成分分析的聚类算法的分割效果.通过提取的非线性特征对数据进行分割有助于识别设备状态的转换,可用于解决一类具有非线性特点的火力发电设备过程状态诊断问题.
Abstract:
To solve the problem of equipment condition diagnosis with nonlinear characteristics, a multivariate time series segmentation algorithm based on kernel principal component analysis(KPCA)and Gath-Geva fuzzy clustering was proposed. The clustering results obtained by the Gath-Geva fuzzy clustering algorithm was used to extract the nonlinear features by the KPCA algorithm and to construct the KPCA analysis model. To improve the segmentation effects on the method, the distance of the clusters in the KPCA model space was used as the criteria for analyzing clusters’ similarity and merging. The experimental results show that the Gath-Geva fuzzy clustering algorithm based on KPCA is used to identify the nonlinear information of the data and analyze the data features more accurately. The segmentation effect is better than that of the clustering algorithm based on principal component analysis. The segmentation of the data by the extracted nonlinear features helps to identify the condition of the equipment, thus it can be used to solve the problem of process condition diagnosis of a class of nonlinear features of thermal power plants.

参考文献/References:

[1] 严英杰, 盛戈皞, 陈玉峰, 等. 基于时间序列分析的输变电设备状态大数据清洗方法[J]. 电力系统自动化, 2015, 39(7): 138-144. DOI:10.7500/AEPS20140111003.
Yan Y J, Sheng G H, Chen Y F, et al. Cleaning method for big data of power transmission and transformation equipment state based on time sequence analysis[J]. Automation of Electric Power Systems, 2015, 39(7): 138-144. DOI:10.7500/AEPS20140111003. (in Chinese)
[2] Aghabozorgi S, Seyed Shirkhorshidi A, Ying Wah T. Time-series clustering—A decade review[J]. Information Systems, 2015, 53: 16-38. DOI:10.1016/j.is.2015.04.007.
[3] Chang S, Wu R, Lv Y, et al. The application of failure prognostic system in state monitoring of power plant generation equipments[C]// IEEE World Automation Congress. Puerto Vallarta, Mexico, 2012:1-4. DOI:10.1097/00008390-199309002-00095.
[4] Esling P, Agon C. Time-series data mining[J]. ACM Computing Surveys, 2012, 45(1): 1-34. DOI:10.1145/2379776.2379788.
[5] Dong R J, Pedrycz W. A granular time series approach to long-term forecasting and trend forecasting[J]. Physica A: Statistical Mechanics and its Applications, 2008, 387(13): 3253-3270. DOI:10.1016/j.physa.2008.01.095.
[6] Abonyi J, Feil B, Nemeth S, et al. Modified Gath-Geva clustering for fuzzy segmentation of multivariate time-series[J]. Fuzzy Sets and Systems, 2005, 149(1): 39-56. DOI:10.1016/j.fss.2004.07.008.
[7] Wang N N, Liu X D, Yin J C. Improved Gath-Geva clustering for fuzzy segmentation of hydrometeorological time series[J]. Stochastic Environmental Research and Risk Assessment, 2012, 26(1): 139-155. DOI:10.1007/s00477-011-0542-0.
[8] 熊辉. 多变量热工系统的自适应控制策略研究[D]. 保定: 华北电力大学, 2012.
  Xiong H. Research on adaptive control strategy of multivariable thermal system[D]. Baoding: North China Electric Power University, 2012.(in Chinese)
[9] Hubert P. The segmentation procedure as a tool for discrete modeling of hydrometeorological regimes[J]. Stochastic Environmental Research and Risk Assessment, 2000, 14(4/5): 297-304. DOI:10.1007/pl00013450.
[10] Himberg J, Korpiaho K, Mannila H, et al. Time series segmentation for context recognition in mobile devices[C]//Proceedings 2001 IEEE International Conference on Data Mining. San Jose, CA, USA, 2001: 203-210. DOI:10.1109/ICDM.2001.989520.
[11] Fezai R, Mansouri M, Taouali O, et al. Online reduced kernel principal component analysis for process monitoring[J]. Journal of Process Control, 2018, 61: 1-11. DOI:10.1016/j.jprocont.2017.10.010.
[12] Dobos L, Abonyi J. On-line detection of homogeneous operation ranges by dynamic principal component analysis based time-series segmentation[J]. Chemical Engineering Science, 2012, 75: 96-105. DOI:10.1016/j.ces.2012.02.022.
[13] Vemulapalli R, Pillai J K, Chellappa R. Kernel learning for extrinsic classification of manifold features[C]// IEEE Conference on Computer Vision and Pattern Recognition. Portland, OR, USA. New York, USA:IEEE, 2013: 1782-1789. DOI:10.1109/CVPR.2013.233.

备注/Memo

备注/Memo:
收稿日期: 2018-12-04.
作者简介: 汪国新(1993—),男,硕士生;郝勇生(联系人),男,博士,副教授,haoys@seu.edu.cn.
基金项目: 国家自然科学基金面上资助项目(51876035).
引用本文: 汪国新,郝勇生,苏志刚.基于KPCA-GG的火力发电设备状态诊断方法[J].东南大学学报(自然科学版),2019,49(3):542-548. DOI:10.3969/j.issn.1001-0505.2019.03.020.
更新日期/Last Update: 2019-05-20