Feature Extraction of Anomaly Electricity Usage Behavior in Residence Using Autoencoder

Abstract

Due to the climate crisis, energy-saving issues and carbon reduction have become the top priority for all countries. Owing to the increasing popularity of advanced metering infrastructure and smart meters, the cost of acquiring data on residential electricity consumption has substantially dropped. This change promotes the analysis of residential electricity consumption, which features both small and complicated consumption behaviors, using machine learning to become an important research topic among various energy saving and carbon reduction measures. The main subtopic of this subject is the identification of abnormal electricity consumption behaviors. At present, anomaly detection is typically realized using models based on low-level features directly collected by sensors and electricity meters. However, due to the significant number of dimensions and a large amount of redundant information in these low-level features, the training efficiency of the model is often low. To overcome this, this study adopts an autoencoder, which is a deep learning technology, to extract the high-level electricity consumption information of residential users to improve the anomaly detection performance of the model. Subsequently, this study trains one-class SVM models for anomaly detection by using the high-level features of five actual residential users to verify the benefits of high-level features.

Advances of Future IoE Wireless Network Technology

“…Another article by Tsai et al [4] describe a method for detecting anomaly electricity usage behavior in residences using an autoencoder. By analyzing electricity usage patterns, the system can identify abnormal behavior that may indicate a potential safety or security issue.…”

Clustering and Investigation of Human Driving Behavior with Shoe Type in Urban Roads using Autoencoder

“…Recurrent Neural Networks (RNN) are used for driving behavior profiling based on the data collected from an accelerometer in a smartphone 11 . In recent years, autoencoder, a deep learning architecture, has gained popularity in various domains due to its ability to learn meaningful representations of high-dimensional data [12][13][14][15][16] . Specifically, autoencoder has been shown to be a useful tool for extracting and clustering meaningful driving behavior with promising results.…”

Detection of Anomalies in the Operation of a Road Lighting System Based on Data from Smart Electricity Meters

“…Random forest, support vector machine, decision tree, naive Bayes, K-nearest neighbor, and neural network algorithms were used to detect anomalies. In [6], the authors presented a method for detecting household anomalous energy consumption based on an autoencoder and SVM. The proposed technology can be integrated into a home energy management system to provide appropriate suggestions for saving energy in a timely manner, owing to its accuracy and speed in detecting abnormal behavior, realizing the concept of edge computing.…”