Real-Time Monitoring and Simulation of Multi-User Electricity Metering Anomaly Data Based on Distributed System

Amidst the rapid development of smart grids and distributed energy systems, the volume and complexity of data within power systems have significantly increased, posing substantial challenges to traditional anomaly detection methods. To overcome these challenges, this study introduces a V-LSTM framew...

Full description

Saved in:
Bibliographic Details
Main Authors: Kai Liu, Xuchao Jia, Junlong Wang, Xun Ma, Jiadong Li
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10930424/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849388494338654208
author Kai Liu
Xuchao Jia
Junlong Wang
Xun Ma
Jiadong Li
author_facet Kai Liu
Xuchao Jia
Junlong Wang
Xun Ma
Jiadong Li
author_sort Kai Liu
collection DOAJ
description Amidst the rapid development of smart grids and distributed energy systems, the volume and complexity of data within power systems have significantly increased, posing substantial challenges to traditional anomaly detection methods. To overcome these challenges, this study introduces a V-LSTM framework, an innovative approach that combines a variational autoencoder (VAE) with a long short-term memory (LSTM) network for anomaly detection in distributed power metering systems. The framework facilitates the extraction of features from multi-source data, generated by users’ electricity consumption behavior, via VAE, while LSTM conducts a meticulous time-series analysis of these features to enable high-precision anomaly detection in complex datasets. VAE proficiently mitigates data complexity, preserving essential anomalous information, whereas LSTM augments the model’s capacity to manage temporal dependencies. In empirical evaluations utilizing both public and proprietary datasets, the V-LSTM framework demonstrates superior performance in accuracy and AUC metrics, decisively surpassing traditional detection approaches. The experimental findings indicate that V-LSTM not only enhances the accuracy and dependability of power anomaly detection but also offers a novel technical trajectory for anomaly monitoring within the smart grid. This research outcome fosters the continued evolution of smart grid technology, providing more scientifically grounded management and decision support for power utilities and energy management entities.
format Article
id doaj-art-32a24b9c2aa84a8a926250a30ffa3fd1
institution Kabale University
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-32a24b9c2aa84a8a926250a30ffa3fd12025-08-20T03:42:18ZengIEEEIEEE Access2169-35362025-01-0113508765088410.1109/ACCESS.2025.355206210930424Real-Time Monitoring and Simulation of Multi-User Electricity Metering Anomaly Data Based on Distributed SystemKai Liu0https://orcid.org/0009-0009-9616-925XXuchao Jia1https://orcid.org/0009-0004-6538-1200Junlong Wang2Xun Ma3Jiadong Li4https://orcid.org/0009-0004-3262-2047Marketing Service Center, State Grid Hebei Electric Power Company Ltd., Shijiazhuang, ChinaMarketing Service Center, State Grid Hebei Electric Power Company Ltd., Shijiazhuang, ChinaMarketing Service Center, State Grid Hebei Electric Power Company Ltd., Shijiazhuang, ChinaMarketing Service Center, State Grid Hebei Electric Power Company Ltd., Shijiazhuang, ChinaMarketing Service Center, State Grid Hebei Electric Power Company Ltd., Shijiazhuang, ChinaAmidst the rapid development of smart grids and distributed energy systems, the volume and complexity of data within power systems have significantly increased, posing substantial challenges to traditional anomaly detection methods. To overcome these challenges, this study introduces a V-LSTM framework, an innovative approach that combines a variational autoencoder (VAE) with a long short-term memory (LSTM) network for anomaly detection in distributed power metering systems. The framework facilitates the extraction of features from multi-source data, generated by users’ electricity consumption behavior, via VAE, while LSTM conducts a meticulous time-series analysis of these features to enable high-precision anomaly detection in complex datasets. VAE proficiently mitigates data complexity, preserving essential anomalous information, whereas LSTM augments the model’s capacity to manage temporal dependencies. In empirical evaluations utilizing both public and proprietary datasets, the V-LSTM framework demonstrates superior performance in accuracy and AUC metrics, decisively surpassing traditional detection approaches. The experimental findings indicate that V-LSTM not only enhances the accuracy and dependability of power anomaly detection but also offers a novel technical trajectory for anomaly monitoring within the smart grid. This research outcome fosters the continued evolution of smart grid technology, providing more scientifically grounded management and decision support for power utilities and energy management entities.https://ieeexplore.ieee.org/document/10930424/Anomaly detectionelectricity meteringVAELSTMdeep learning
spellingShingle Kai Liu
Xuchao Jia
Junlong Wang
Xun Ma
Jiadong Li
Real-Time Monitoring and Simulation of Multi-User Electricity Metering Anomaly Data Based on Distributed System
IEEE Access
Anomaly detection
electricity metering
VAE
LSTM
deep learning
title Real-Time Monitoring and Simulation of Multi-User Electricity Metering Anomaly Data Based on Distributed System
title_full Real-Time Monitoring and Simulation of Multi-User Electricity Metering Anomaly Data Based on Distributed System
title_fullStr Real-Time Monitoring and Simulation of Multi-User Electricity Metering Anomaly Data Based on Distributed System
title_full_unstemmed Real-Time Monitoring and Simulation of Multi-User Electricity Metering Anomaly Data Based on Distributed System
title_short Real-Time Monitoring and Simulation of Multi-User Electricity Metering Anomaly Data Based on Distributed System
title_sort real time monitoring and simulation of multi user electricity metering anomaly data based on distributed system
topic Anomaly detection
electricity metering
VAE
LSTM
deep learning
url https://ieeexplore.ieee.org/document/10930424/
work_keys_str_mv AT kailiu realtimemonitoringandsimulationofmultiuserelectricitymeteringanomalydatabasedondistributedsystem
AT xuchaojia realtimemonitoringandsimulationofmultiuserelectricitymeteringanomalydatabasedondistributedsystem
AT junlongwang realtimemonitoringandsimulationofmultiuserelectricitymeteringanomalydatabasedondistributedsystem
AT xunma realtimemonitoringandsimulationofmultiuserelectricitymeteringanomalydatabasedondistributedsystem
AT jiadongli realtimemonitoringandsimulationofmultiuserelectricitymeteringanomalydatabasedondistributedsystem