A Hybrid Model for Anomalies Detection in AMI System Combining K-means Clustering and Deep Neural Network

Abstract

Recently, the radical digital transformation has deeply affected the traditional electricity grid and transformed it into an intelligent network (smart grid). This mutation is based on the progressive development of advanced technologies: advanced metering infrastructure (AMI) and smart meter which play a crucial role in the development of smart grid. AMI technologies have a promising potential in terms of improvement in energy efficiency, better demand management, and reduction in electricity costs. However the possibility of hacking smart meters and electricity theft is still among the most significant challenges facing electricity companies. In this regard, we propose a hybrid approach to detect anomalies associated with electricity theft in the AMI system, based on a combination of two robust machine learning algorithms; K-means and Deep Neural Network (DNN). K-means unsupervised machine learning algorithm is used to identify groups of customers with similar electricity consumption patterns to understand different types of normal behavior. DNN algorithm is used to build an accurate anomaly detection model capable of detecting changes or anomalies in usage behavior and deciding whether the customer has a normal or malicious consumption behavior. The proposed model is constructed and evaluated based on a real dataset from the Irish Smart Energy Trials. The results show a high performance of the proposed model compared to the models mentioned in the literature.

A Combined Deep Learning and Ensemble Learning Methodology to Avoid Electricity Theft in Smart Grids

“…Thus, machine learning solutions are further categorized into the supervised and unsupervised learning methods. Unsupervised learning methods have proposed clustering-based solutions to group the similar instances into one cluster [32,33], members of which each have a high false positive rate (FPR). In this paper, a unique supervised learning-based solution, namely, the LSTM-UNet-Adaboost, is proposed to perform the binary classification using the data from on-site inspections.…”

“…Tianyu et al [38] made a semi-supervised deep learning model, known as the multitask feature extracting fraud detector (MFEFD), for ETD: both the supervised and unsupervised learning procedures are combined to capture important features from the labeled and unlabeled data. Maamar et al [32] offered a hybrid model that utilizes the k-means clustering procedure and deep neural networks (DNN) for ETD in the AMI system, where k-means is utilized to gather consumers having the same electricity consumption behaviors, and DNN is used to detect anomalies in the electricity consumption behaviors of the consumers. Ghasemi et al [26] proposed a solution comprising of a probabilistic neural network and mathematical model, named the PNN-Levenberg-Marquardt, for the identification of two types of illegal consumers using the observer's meters.…”

Network Traffic Anomaly Detection Based on ML-ESN for Power Metering System

“…Among them, researchers at home and abroad have applied network intrusion detection technology to anomaly detection of AMI network flow and proposed a variety of anomaly detection and analysis models, such as deep neural network [1], Markov [4], density statistics [5], BP neural network [6], attack graph-based information fusion [7], and principle component analysis [8].…”

Density Peak Clustering Algorithm Considering Topological Features

“…Based on previous studies, we propose a clustering algorithm that adds the topological characteristics of original data on the basis of the CFSFDP algorithm. Our experimental results show that the clustering algorithm with topological features significantly improves the clustering effect and proves that the addition of topological features is effective and feasible.Electronics 2020, 9, 459 2 of 16The K-means algorithm is an algorithm that adopts the alternative minimization method to solve non-convex optimization problems [11,12] and it is a representative of the prototype-based clustering method of objective functions. It divides a given data set into K clusters designated by users and has a high execution efficiency.…”

“…The K-means algorithm is an algorithm that adopts the alternative minimization method to solve non-convex optimization problems [11,12] and it is a representative of the prototype-based clustering method of objective functions. It divides a given data set into K clusters designated by users and has a high execution efficiency.…”