MIDAS: Detection of Non-technical Losses in Electrical Consumption Using Neural Networks and Statistical Techniques

Suspicious electric consumption detection based on multi-profiling using live machine learning

“…Most recent approaches are based on load profiling, becoming possible through automated meter reading. Monedero [17] et al, propose to apply data mining techniques for detection of non-technical losses and present two methodologies, one based on neural networks and one on statistical techniques. They classify customers in two groups, i) likely to be affected and ii) not affected by non-technical losses.…”

Detection of non-technical losses in smart meter data based on load curve profiling and time series analysis

“…To begin with, several contributions have gravitated on the use of machine learning models over supervised datasets, such as Support Vector Machines [13,14,15,16,17,18], Neural Networks [19,20,21], Extreme Learning Machines [22], Path Forests [23,24], Decision Trees [25,26,27], model ensembles [28], and statistical methods [29,30]. However, all such previous work builds upon the assumption that supervised datasets capture the entire casuistry of symptomatic anomalies of interest for fraud detection and/or electricity theft, which not only unrealistic in practice but also yields highly imbalanced datasets that subsequently jeopardize the model learning process.…”

Fraud Detection in Electric Power Distribution Networks using an Ann-Based Knowledge-Discovery Process

“…At that study, a knowledge-discovery methodology with artificial intelligence for data preprocessing and mining was successfully applied in different scenarios. Monedero et al [3] developed an approach to identify non-technical losses in electric power distribution networks using artificial neural networks applied to the city of Seville (Spain), outperforming by over 50% precision compared to the previous methodologies.…”