Improving clustering based anomaly detection with concave hull: An application in fault diagnosis of wind turbines

Along with the rapid growth of the system complexity, the capability of self-diagnosis is desired by monitoring complex industrial systems to reduce the unplanned system downtimes. By applying data driven analysis methods such as clustering algorithms on the process data of industrial systems, the health status of systems can be deduced and the anomalous statuses can be automatically detected. The accuracy of clustering based anomaly detection using cluster centers is highly dependent on the geometry of the given data set. By a data set with unsymmetrical and concave boundary, using cluster centers as reference to measure the similarity between new observations and clusters normally leads to a high false alarm rate. This paper presented an approach to improve clustering based anomaly detection by building concave hulls for each cluster. For this purpose, a new algorithm for generating n-dimensional concave hulls is developed. The effectiveness of this approach is evaluated with real world data collected from wind turbines.

[1]  Peng Li,et al.  Data Driven Condition Monitoring of Wind Power Plants Using Cluster Analysis , 2015, 2015 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery.

[2]  Gary B. Wills,et al.  Unsupervised Clustering Approach for Network Anomaly Detection , 2012, NDT.

[3]  Asdrúbal López Chau,et al.  Convex-Concave Hull for Classification with Support Vector Machine , 2012, 2012 IEEE 12th International Conference on Data Mining Workshops.

[4]  Se-Jong Oh,et al.  A New Concave Hull Algorithm and Concaveness Measure for n-dimensional Datasets , 2012, J. Inf. Sci. Eng..

[5]  P. Oh,et al.  A n-dimensional convex hull approach for fault detection and mitigation for high degree of freedom robots humanoid robots , 2012, 2012 12th International Conference on Control, Automation and Systems.

[6]  VARUN CHANDOLA,et al.  Anomaly detection: A survey , 2009, CSUR.

[7]  Li Guo,et al.  An automatic approach to extract the formats of network and security log messages , 2015, MILCOM 2015 - 2015 IEEE Military Communications Conference.

[8]  David P. Dobkin,et al.  The quickhull algorithm for convex hulls , 1996, TOMS.

[9]  Juntae Kim,et al.  The Anomaly Detection by Using DBSCAN Clustering with Multiple Parameters , 2011, 2011 International Conference on Information Science and Applications.

[10]  Mete Celik,et al.  Anomaly detection in temperature data using DBSCAN algorithm , 2011, 2011 International Symposium on Innovations in Intelligent Systems and Applications.

[11]  Jeff Jones.  Material Representation of Area and Shape: Convex Hull, Concave Hull and Skeleton , 2015 .

[12]  Maribel Yasmina Santos,et al.  Concave hull: A k-nearest neighbours approach for the computation of the region occupied by a set of points , 2007, GRAPP.

[13]  Ali Mohades,et al.  Alpha Convex Hull, a Generalization of Convex Hull , 2013, ArXiv.