Abstract
Abstract With the digitalization of electricity meters many previously solved security problems, such as electricity theft, are reintroduced as IT related challenges which require modern detection schemes based on data analysis, machine learning and forecasting. Here, we demonstrate a multidimensional anomaly detection approach for the early detection of tampered with electricity meters by comparing a set of multiple energy demand time series. Our method can complement and enhance existing monitoring systems which usually only analyze a single time series. We aim to detect electricity theft, which leads to noticeable outliers in our work. We present three data preprocessing methods to produce outliers in case of energy theft and highlight the requirements and fine-tuning mechanisms for the aggregation and comparison of multiple data sources. We show that our metric is robust against multiple manipulated data sources, which is a concrete improvement to alternative outlier preserving concepts to aggregate multiple data sources. With detection rates better than 90%, we demonstrate the effectiveness of using several data sources simultaneously, that, when used individually, provide little value in anomaly detection. Furthermore, we show that we can use different households as comparable data sources, without clustering the households according to their similarity first.
Highlights
The success of renewable energy usage is fueling the power grids most significant transformation seen in decades, from a centrally controlled electricity supply towards an intelligent, decentralized power supply
We summarize the relationship between these topics, to smart grid security and anomaly detection
The entropy-inspired metric in combination with time series prediction, results in acceptable detection rates mostly over 90%, which can compete with the other function, it only performs better if we have several outliers in the distribution – which means that e.g. the current day and day before show energy theft
Summary
The success of renewable energy usage is fueling the power grids most significant transformation seen in decades, from a centrally controlled electricity supply towards an intelligent, decentralized power supply. Illera et al [1], demonstrated that smart meters installed in Spain used strong symmetric encryption but stored a static encryption key in a plain text file, which allows adversaries to artificially manipulate and tamper with the data and measurements of a smart meters communication channel. For this reason, there is a high interest in utilizing the fine-grained data and recent advances in. We introduce and validate requirements on the input data and highlight cases in which our entropyinspired metric leads to a concrete improvement compared to alternative methods aggregating different data sources. We compare our approach with two alternative anomaly detection schemes based on Naive Bayes and XMR charts
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have