The Turbine Machine Fault Prediction Based on Kernel Principal Component Analysis

Lu Hui Lin; Jie Ma; Xiao Li Xu

doi:10.4028/www.scientific.net/AMR.383-390.4787

Paper Titles

Analysis and Optimization Design of Self Magnetic-Suspension Permanent Magnet Linear Synchronous Motor
p.4762

An Improved Particle Swarm Optimization Approach for Temperature Control in HVAC for the Purpose of Energy Saving
p.4768

Analysis of Precision Reliability for CX Series of Five-Axis Turning-Milling Center
p.4775

Orthogonal Experimental Design Applied in Analyzing Factors Influencing on Heat Transfer Performance of Engine Cylinder Head
p.4783

The Turbine Machine Fault Prediction Based on Kernel Principal Component Analysis
p.4787

Research on SSO Caused by HVDC Based on Eigenvalue Analysis
p.4792

Vehicle Model Recognition Based on Fuzzy Pattern Recognition Method
p.4799

Structure and Finite Element Analysis for a Permanent Magnet Bias Axial Magnetic Bearing
p.4803

Glare Prediction Method Based on SPEOS Simulation for Civil Airplane Flight Deck
p.4810

HomeAdvanced Materials ResearchAdvanced Materials Research Vols. 383-390The Turbine Machine Fault Prediction Based on...

The Turbine Machine Fault Prediction Based on Kernel Principal Component Analysis

Abstract:

Kernel principal component analysis (KPCA) is presented and is applied to predict the huge electro-mechanical system fault. Take the gas turbine set of Beijing Yanshan Petrochemical Refinery as the research object. KPCA uses the historical normal data of vibration intensity value to establish a prediction system. And then it is used to forecast the collected data for judging whether the turbine set is in normal. The simulation experiment result indicates the effectiveness of the method and the running state can be judged as normal or not from the result. And the experiment also shows KPCA can obtain a satisfactory prediction result.

You might also be interested in these eBooks

Manufacturing Science and Technology, ICMST2011

View Preview

Info:

Periodical:

Advanced Materials Research (Volumes 383-390)

Pages:

4787-4791

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.383-390.4787

Citation:

Cite this paper

Online since:

November 2011

Authors:

Lu Hui Lin, Jie Ma, Xiao Li Xu

Keywords:

Fault Prediction, KPCA, Turbine Machine

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] Xu Xiaoli, Xu Yong, Wang Xinyi, Research for trend indicating technology of machine working state, Journal of INDUSTRIAL INSTRUMENTATION & AUTOMATION, vol. 3, pp.9-12, (1998).

Google Scholar

[2] G.K. Singh and Sad Ahmed Saleh. AI Kazzaz, Induction machine drive condition monitoring and diagnostic research-a survey, Journal of Electric Power Systems Research, vol. 64, No. 2, p.145~158, (2003).

DOI: 10.1016/s0378-7796(02)00172-4

Google Scholar

[3] Norden E Huang. A New method for nonlinear and non-stationary time series analysis and its application for civil infrastructure health monitoring. NASA Goddard Space Flight Center, Greenbelt, Maryland 20771, USA.

Google Scholar

[4] Scholkopf B, Smola A, Mult]ler K R., Nolinear component analysis as a kernel eigenvalue problem, Journal of Neural Computation, vol. 10, No. 5, p.1299~1319, (1998).

DOI: 10.1162/089976698300017467

Google Scholar

[5] Fan Yu-gang, Li Ping, Song Zhi-huan, KPCA based on feature samples for fault detection, Journal of Control and Decision, vol. 20, No. 12, p.1415~1422, (2005).

Google Scholar

[6] Kourti T, MacGregor J F., Multivariate SPC methods for process and product monitoring, J. Qual Technol, vol. 28, p.409~421, (1996).

DOI: 10.1080/00224065.1996.11979699

Google Scholar

[7] Jackson J E., A user's to principle components, New York: Wiley-Inter-Science, (1991).

Google Scholar