A New SVM-Based Fraud Detection Model for AMI

This paper presents a new strategy for fraud detection in Advanced Metering Infrastructure (AMI) based on the analysis of disturbances in the pattern consumption of end-customers. The proposed strategy is based on the use of SVM (Supported Vector Machine). SVM requires labeled training data in order to define a classification function. The need of labeled data is a serious limitation for practical implementation of fraud detection systems in AMI. To work around this problem, we propose a new strategy for training SVM classifiers that requires only normal consumption patterns in the training phase. The anomalous consumption is generated by simulating attacks on the normal consumption patterns.

[1]  Yang Xiao,et al.  Non-repudiation in neighborhood area networks for smart grid , 2013, IEEE Communications Magazine.

[2]  Saman A. Zonouz,et al.  A Multi-Sensor Energy Theft Detection Framework for Advanced Metering Infrastructures , 2013, IEEE Journal on Selected Areas in Communications.

[3]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[4]  A. N. de Souza,et al.  Detection and Identification of Abnormalities in Customer Consumptions in Power Distribution Systems , 2011, IEEE Transactions on Power Delivery.

[5]  Yang Xiao,et al.  Exploring Malicious Meter Inspection in Neighborhood Area Smart Grids , 2013, IEEE Transactions on Smart Grid.

[6]  Sieh Kiong Tiong,et al.  Nontechnical Loss Detection for Metered Customers in Power Utility Using Support Vector Machines , 2010, IEEE Transactions on Power Delivery.

[7]  Jun Luo,et al.  Energy-theft detection issues for advanced metering infrastructure in smart grid , 2014, Tsinghua Science and Technology.

[8]  Pan Li,et al.  Privacy-preserving energy theft detection in smart grids , 2012, 2012 9th Annual IEEE Communications Society Conference on Sensor, Mesh and Ad Hoc Communications and Networks (SECON).

[9]  Yang Xiao,et al.  Building Accountable Smart Grids in Neighborhood Area Networks , 2011, 2011 IEEE Global Telecommunications Conference - GLOBECOM 2011.

[10]  Florian Skopik,et al.  Attack Vectors to Metering Data in Smart Grids under Security Constraints , 2012, 2012 IEEE 36th Annual Computer Software and Applications Conference Workshops.

[11]  Lingfeng Wang,et al.  Support vector machine based data classification for detection of electricity theft , 2011, 2011 IEEE/PES Power Systems Conference and Exposition.

[12]  Pedro Antmann,et al.  Reducing Technical and Non-Technical Losses in the Power Sector , 2009 .

[13]  Lingfeng Wang,et al.  High performance computing for detection of electricity theft , 2013 .

[14]  Nadeem Javaid,et al.  Minimizing Electricity Theft Using Smart Meters in AMI , 2012, 2012 Seventh International Conference on P2P, Parallel, Grid, Cloud and Internet Computing.