Abnormal behaviour analysis algorithm for electricity consumption based on density clustering

How to effectively detect abnormal electricity consumption behaviour from a large-scale electrical load data is very important to smart grid. An abnormal electricity consumption analysis method based on density clustering is proposed. First, the similar users located in fix area are clustered in accordance with the electricity consumption characteristics. Then, on the basis of electricity consumption data sequence, the outlier electricity consumption for the similar users are obtained with density clustering. The matching degrees of these outliers can be calculated based on the similar user electricity consumption model and historical electricity consumption model. Finally, according to the threshold value, the abnormal electricity consumption can be discriminated with the comprehensive support degree. The simulation results show that this method can effectively identify the abnormal electricity consumption behaviour.

[1]  Pan Gao,et al.  A Novel Fault Early Warning Model Based on Fault Gene Table for Smart Distribution Grids , 2017 .

[2]  Canbing Li,et al.  Flexible Voltage Control Strategy Considering Distributed Energy Storages for DC Distribution Network , 2019, IEEE Transactions on Smart Grid.

[3]  KokSheik Wong,et al.  Detection of energy theft and defective smart meters in smart grids using linear regression , 2017 .

[4]  Azadeh Faroughi,et al.  CANF: Clustering and anomaly detection method using nearest and farthest neighbor , 2018, Future Gener. Comput. Syst..

[5]  Naixue Xiong,et al.  Improved clustering algorithm based on energy consumption in wireless sensor networks , 2017, IET Networks.

[6]  Shanlin Yang,et al.  Outlier Data Treatment Methods Toward Smart Grid Applications , 2018, IEEE Access.

[7]  Victor C. M. Leung,et al.  Electricity Theft Detection in AMI Using Customers’ Consumption Patterns , 2016, IEEE Transactions on Smart Grid.

[8]  Christian Rehtanz,et al.  Optimal energy management for the residential MES , 2019, IET Generation, Transmission & Distribution.

[9]  Robin Girard,et al.  Long Term Forecast of Local Electrical Demand and Evaluation of Future Impacts on the Electricity Distribution Network , 2017 .

[10]  Meera Dhabu,et al.  Motif-based association rule mining and clustering technique for determining energy usage patterns for smart meter data , 2019, Sustainable Cities and Society.

[11]  G. Kariniotakis,et al.  Probabilistic day-ahead forecasting of household electricity demand , 2017 .