A Novel Ensemble Clustering for Operational Transients Classification with Application to a Nuclear Power Plant Turbine

Sameer Al-Dahidi, Francesco Di Maio, Piero Baraldi, Enrico Zio, and Redouane Seraoui
Publication Target: 
IJPHM
Publication Issue: 
Special Issue Nuclear Energy PHM
Submission Type: 
Full Paper
Supporting Agencies (optional): 
The European Union Project INNovation through Human Factors in risk analysis and management (INNHF, www.innhf.eu) funded by the 7th framework program FP7-PEOPLE-2011- Initial Training Network: Marie-Curie Action.
AttachmentSizeTimestamp
ijphm_15_001.pdf1.8 MBMay 27, 2015 - 9:13am

The objective of the present work is to develop a novel approach for combining in an ensemble multiple base clusterings of operational transients of industrial equipment, when the number of clusters in the final consensus clustering is unknown. A measure of pairwise similarity is used to quantify the co-association matrix that describes the similarity among the different base clusterings. Then, a Spectral Clustering technique of literature, embedding the unsupervised K-Means algorithm, is applied to the co-association matrix for finding the optimum number of clusters of the final consensus clustering, based on Silhouette validity index calculation. The proposed approach is developed with reference to an artificial case study, properly designed to mimic the signal trend behavior of a Nuclear Power Plant (NPP) turbine during shut-down. The results of the artificial case have been compared with those achieved by a state-of-art approach, known as Cluster-based Similarity Partitioning and Serial Graph Partitioning and Fill-reducing Matrix Ordering Algorithms (CSPA-METIS). The comparison shows that the proposed approach is able to identify a final consensus clustering that classifies the transients with better accuracy and robustness compared to the CSPA-METIS approach. The approach is, then, validated on an industrial case concerning 149 shut-down transients of a NPP turbine.

Publication Year: 
2015
Publication Volume: 
6
Publication Control Number: 
001
Page Count: 
21
Submission Keywords: 
unsupervised learning
Ensemble Clustering
Final Consensus Clustering
Spectral Clustering
Operational Transients
Nuclear Power Plant (NPP) turbine shut-down.
Submission Topic Areas: 
Data-driven methods for fault detection, diagnosis, and prognosis
Industrial applications
Submitted by: 
  
 
 
 

follow us

PHM Society on Facebook Follow PHM Society on Twitter PHM Society on LinkedIn PHM Society RSS News Feed