Data for: Using Multiple Data Sources to Detect Manipulated Electricity Meter by an Entropy-Inspired Metric

Published: 11 January 2020| Version 1 | DOI: 10.17632/xczmmd2g64.1
Contributors:
Denis Hock, Martin Kappes, Bogdan Ghita

Description

This GNU R package contains sample data of the ECO data as well as all methods introduced in this paper, including methods to derive the 'activity'-features, generate a matrix with multiple datasources and compute the entropy. An example usage is described in the documentation. Package: DemandAnomalyDetection Title: Energy Demand Anomaly Detection Version: 0.1-1 Maintainer: Denis Hock <dehock@fb2.fra-uas.de> Description: Anomaly Detection according to the article - Using Multiple Data Sources to Detect Manipulated Electricity Meter by an Entropy-Inspired Metric. The input data is a time series of power (W). Using this package, we can compute higher level features to easily detect energy theft. It is also possible to compute the entropy in order to compare several data sources.

Files

Categories

Detection System, Theft, Smart Grid

Licence