Deep Learning Detection of Inaccurate Smart Electricity Meters: A Case Study

TL;DR

This paper presents a deep learning approach combining LSTM and CNN to detect inaccurate smart electricity meters by analyzing discrepancies between predicted and observed usage trajectories, aiming to optimize replacement strategies.

Contribution

A novel deep learning method integrating LSTM and CNN for accurate detection of inaccurate smart meters based on usage trajectory analysis.

Findings

High accuracy in detecting inaccurate meters demonstrated in case study

Method effectively distinguishes between accurate and faulty meters

Potential to reduce unnecessary replacements and extend smart meter lifespan

Abstract

Detecting inaccurate smart meters and targeting them for replacement can save significant resources. For this purpose, a novel deep-learning method was developed based on long short-term memory (LSTM) and a modified convolutional neural network (CNN) to predict electricity usage trajectories based on historical data. From the significant difference between the predicted trajectory and the observed one, the meters that cannot measure electricity accurately are located. In a case study, a proof of principle was demonstrated in detecting inaccurate meters with high accuracy for practical usage to prevent unnecessary replacement and increase the service life span of smart meters.