Electricity Fraud Research Articles

Fraud Detection of the Electricity Consumption by combining Deep Learning and Statistical Methods

An important issue for the electricity distribution companies is the non-technical loss (NTL), also known as electricity fraud. This issue has a significant impact on the economies of all countries in the world. In this context, we studied the problem of the imbalance between the electrical energy invoiced and the electrical energy supplied within the Algerian economic sector. This article presents an approach to detecting electrical fraud using a combination of Long Short-Term Memory (LSTM) and robust Exponential and Holt-Winters Smoothing (EHWS) methods in order to enhance the accuracy and efficacy of fraud detection mechanisms. The proposed approach investigates the fraudulent behaviour of electricity consumers and unfolds in several key phases. In the first step, monthly consumption forecasts are made and the model with the most accurate results is selected. Then, the phase of detection of anomalies in economic meters and detection of cases of fraud by economic customers begins. This phase relies on the robust exponential and Holt-Winters Smoothing methods for uncovering irregular patterns indicative of potential fraud instances. The proposed model was trained and evaluated, and several experiments were carried out using a large dataset of real users from the economic sector. The dataset comprised approximately 2,000 customers and encompassed 14 years of monthly electricity usage in Bejaia, Algeria. The results of the experiments demonstrate promising performance, underscoring the efficacy of our proposed solution in effectively detecting instances of fraud, leading us to conclude that this proposition is robust and can help improve the accuracy of locating abnormal consumer behaviour and increase the company's profits.

An evaluation of nature-inspired optimization algorithms and machine learning classifiers for electricity fraud prediction

<span>This study evaluated the nature-inspired optimization algorithms to improve classification involving imbalanced class problems. The particle swarm optimization (PSO) and grey wolf optimizer (GWO) were used to adaptively balance the distribution and then four supervised machine learning classifiers artificial neural network (ANN), support vector machine (SVM), extreme gradient-boosted tree (XGBoost), and random forest (RF) were applied to maximize the classification performance for electricity fraud prediction. The imbalance data was balanced using random undersampling (RUS) and two nature-inspired algorithm techniques (PSO and GWO). Results showed that for the data balanced using random undersampling, ANN (Sentest = 50.31%), and XGBoost (Sentest = 66.32%) has better sensitivity than SVM (Sentest = 23.61%), while RF exhibits overfitting (Sentrain = 100%, Sentest = 71.25%). The classification performance of RF model hybrid with PSO improved tremendously (AccTest = 96.98%, Sentest = 94.87%, Spectest = 99.16%, Pretest = 99.14%, F1 Score = 96.96%, and area under the curve (AUC) = 0.989). This was closely followed by hybrid of XGBoost with PSO. Moreover, RF and XGBoost hybrid with GWO also showed an improvement and promising results. This study has showed that nature-inspired optimization algorithms (PSO and GWO) are effective methods in addressing imbalanced dataset.</span>

Deep Learning Models for Efficient Detection of Electricity Fraud in Smart Grids

—Electricity theft is one of the biggest problems facing energy companies. These companies eventually claimed that the conventional methods for preventing electricity fraud were insufficient, which led to the creation of systems based on artificial intelligence to identify thieves among electricity consumers. In this paper, we propose a system based on deep learning to identify customers who have engaged in fraudulent activity on smart grids. We selected one-dimensional (1D) and twodimensional (2D) convolutional neural network models from deep learning models to achieve our objective. Also, we proposed a new method to fill in missing values in the data set. Our findings show that our models enhance the performance of systems that identify electricity thieves.

A two stage approach to electricity theft detection in AMI using deep learning

Electricity fraud as a noteworthy portion of non-technical losses has continuously been a worldwide concern. On the other hand, once a fraud is detected, on-site inspection is necessary for final verification. Since inspecting all metering data is laborious and expensive, utilities constantly try to narrow the inspection to instances with a higher odds of fraud. The rise of smart meters has introduced AI-based methods to reduce the inspection's scope by leveraging consumption patterns. However, their performances are limited due to the insufficiency of malicious samples, known as data imbalance. In the following investigation, a two-stage deep-learning-based model is provided for detecting energy fraud using metering data. first, a deep architecture is developed to model suspicious behaviour of fraudulent customers; which is used to handle the problem of data imbalance by predicting possible theft scenarios for normal consumers. then, A DNN network is developed to distinguishes normal and suspicious consumers. Consideration use of clustering and deep learning, as well as concurrent use of consumption and livelihood data, has enabled the algorithm to understand customer's behaviour and not fall against non-malicious changes in consumption patterns. Evaluation of the following method on a repository with 6000 real-world metering data demonstrates its high performance.

Review of the Data-Driven Methods for Electricity Fraud Detection in Smart Metering Systems

In smart grids, homes are equipped with smart meters (SMs) to monitor electricity consumption and report fine-grained readings to electric utility companies for billing and energy management. However, malicious consumers tamper with their SMs to report low readings to reduce their bills. This problem, known as electricity fraud, causes tremendous financial losses to electric utility companies worldwide and threatens the power grid’s stability. To detect electricity fraud, several methods have been proposed in the literature. Among the existing methods, the data-driven methods achieve state-of-art performance. Therefore, in this paper, we study the main existing data-driven electricity fraud detection methods, with emphasis on their pros and cons. We study supervised methods, including wide and deep neural networks and multi-data-source deep learning models, and unsupervised methods, including clustering. Then, we investigate how to preserve the consumers’ privacy, using encryption and federated learning, while enabling electricity fraud detection because it has been shown that fine-grained readings can reveal sensitive information about the consumers’ activities. After that, we investigate how to design robust electricity fraud detectors against adversarial attacks using ensemble learning and model distillation because they enable malicious consumers to evade detection while stealing electricity. Finally, we provide a comprehensive comparison of the existing works, followed by our recommendations for future research directions to enhance electricity fraud detection.

Predictive Data Analytics for Electricity Fraud Detection Using Tuned CNN Ensembler in Smart Grid

In the smart grid (SG), user consumption data are increasing very rapidly. Some users consume electricity legally, while others steal it. Electricity theft causes significant damage to power grids, affects power supply efficiency, and reduces utility revenues. This study helps utilities reduce the problems of electricity theft, inefficient electricity monitoring, and abnormal electricity consumption in smart grids. To this end, an electricity theft dataset from the state grid corporation of China (SGCC) is employed and this study develops a novel model, a mixture of convolutional neural network and gated recurrent unit (CNN-GRU), for automatic power theft detection. Moreover, the hyperparameters of the proposed model are tuned using a meta-heuristic method, the cuckoo search (CS) algorithm. The class imbalance problem is solved using the synthetic minority oversampling technique (SMOTE). The clean data are trained and then tested with the proposed classification. Extensive simulations are performed based on real energy consumption data. The simulated results show that the proposed theft detection model (CNN-GRU-CS) solved the theft classification problem better than other approaches in terms of effectiveness and accuracy by 10% on average. The calculated accuracy of the proposed method is 92% and the precision is 94%.

A Two-Stage Model to Detect Electricity Fraud in The Distribution Network Using Deep Learning

A Two-Stage Model to Detect Electricity Fraud in The Distribution Network Using Deep Learning

Nontechnical Loss Detection With Duffing–Holmes Self-Synchronization Dynamic Errors and 1D CNN-Based Multilayer Classifier in a Smart Grid

The combined Duffing–Holmes (D–H)–based quantizer and one-dimensional (1D) convolutional neural network (CNN)-based multilayer classifier were applied to perform the nontechnical loss (NTL’s) (electricity fraud) feature quantification, feature extraction, and classification tasks to analyze electricity consumption data and to identify either normal or abnormal consumption patterns for NTL detection. The metering data is gathered every 15 minutes and a 3-hour screening window is used to distinguish between the normal conditions and likely NTL events or power outage events. The D–H-based quantizer in the feature quantification layer may quantify the different levels among three events for preliminary screening differences using D–H self-synchronization dynamic errors. In the feature extraction layer, two 1D convolutional-pooling processes are used to extract 1D key feature signals to enhance the distinguished levels for further classification applications. The gray relational analysis (GRA)-based multilayer network is trained as a classifier. In the classification layer, to identify electricity fraud events. The proposed method is verified and validated using simulation data and the electricity fraud attack model. The correlation coefficient and unitary averaged changed intensity index are applied in correlation analysis to discover apparent abnormality between historical consumption and metering consumption patterns within the short-time monitoring. The D-H-based quantizer and 1D CNN-based classifier then work together to accomplish classification tasks on the time-series metering data. The experimental results show that the suggested classifier model demonstrates promising performance and efficiency compared to the traditional multilayer classifier in feature extraction, training, and recall processing, and accurate classification.

Machine learning in electricity fraud detection in smart grids with multivariate Gaussian distribution

Abstract Smart meters allow electricity consumption readings at a high time resolution generating time series that can be investigated to extract valuable insights and detect frauds. Using a dataset with recordings from Chinese consumers, we propose an exploratory data analysis and processing to train several classifiers and assess the results. Good results are obtained with ensemble classifiers such as Random Forest (RF), eXtreme Gradient Boosting (XGB) and Multi-Layer Perceptron (MLP) with two layers and a relatively small number of neurons. Real-consumption dataset daily recorded in China consisting of over 42,000 consumers and over 1,000 days is processed with machine learning ML algorithms or classifiers to distinguish between normal and suspicious consumers. In this paper, we will compare a simple feature engineering method that consists in aggregating the data, calculating distances and density function with no feature engineering, proving that the first approach enhances the results and reduces the utility companies’ costs related to on-site inspections. The results are compared with AUC score and ROC curves as the input data is highly skewed.

A framework to select a classification algorithm in electricity fraud detection

In the electrical domain, a non-technical loss often refers to energy used but not paid for by a consumer. The identification and detection of this loss is important as the financial loss by the electricity supplier has a negative impact on revenue. Several statistical and machine learning classification algorithms have been developed to identify customers who use energy without paying. These algorithms are generally assessed and compared using results from a confusion matrix. We propose that the data for the performance metrics from the confusion matrix be resampled to improve the comparison methods of the algorithms. We use the results from three classification algorithms, namely a support vector machine, k-nearest neighbour and naïve Bayes procedure, to demonstrate how the methodology identifies the best classifier. The case study is of electrical consumption data for a large municipality in South Africa.
 Significance:
 
 The methodology provides data analysts with a procedure for analysing electricity consumption in an attempt to identify abnormal usage.
 The resampling procedure provides a method for assessing performance measures in fraud detection systems.
 The results show that no single metric is best, and that the selected metric is dependent on the objective of the analysis.

An Efficient Boosted C5.0 Decision-Tree-Based Classification Approach for Detecting Non-Technical Losses in Power Utilities

Electricity fraud in billing are the primary concerns for Distribution System Operators (DSO). It is estimated that billions of dollars are wasted annually due to these illegal activities. DSOs around the world, especially in underdeveloped countries, still utilize conventional time consuming and inefficient methods for Non-Technical Loss (NTL) detection. This research work attempts to solve the mentioned problem by developing an efficient energy theft detection model in order to identify the fraudster customers in a power distribution system. The key motivation for the present study is to assist the DSOs in their fight against energy theft. The proposed computational model initially utilizes a set of distinct features extracted from the monthly consumers’ consumption data, obtained from Multan Electric Power Company (MEPCO) Pakistan, to segregate the honest and the fraudulent customers. The Pearson’s chi-square feature selection algorithm is adopted to select the most relevant features among the extracted ones. Finally, the Boosted C5.0 Decision Tree (DT) algorithm is used to classify the honest and the fraudster consumers based on the outcomes of the selected features. To validate the superiority of the proposed NTL detection approach, its performance is matched with that of few state-of-the-art machine learning algorithms (one of most exciting recent technologies in Artificial Intelligence), like Random Forest (RF), Support Vector Machine (SVM), Artificial Neural Network (ANN) and Extreme Gradient Bossting (XGBoost). The proposed NTL detection method provides an accuracy of 94.6%, Sensitivity of 78.1%, Specificity of 98.2%, F1 score 84.9% and Precision of 93.2% which are significantly higher than that of the same for the above-mentioned algorithms.

A Cryptographic-Based Approach For Electricity Theft Detection in Smart Grid

In order to strengthen their security issues, electrical companies devote particular efforts to developing and enhancing their fraud detection techniques that cope with the information and communication technologies integration in smart grid fields. Having been treated earlier by several researchers, various detection schemes adapted from attack models that benefit from the smart grid topologies weaknesses, aiming primarily to the identification of suspicious incoming hazards. Wireless meshes have been extensively used in smart grid communication architectures due to their facility, lightness of conception and low cost installation; however, the communicated packets are still exposed to be intercepted maliciously in order either to falsify pertinent information like the smart meter readings, or to inject false data instead, aiming at electricity theft during the communication phase. For this reason, this paper initiates a novel method based on RSA cryptographic algorithm to detect electricity fraud in smart grid. This new method consists of generating two different cryptograms of one electricity measurement before sending, after which the recipient is used to find the same value after decrypting the two cyphers in a normal case. Otherwise, a fraudulent manipulation could occur during the transmission stage. The presented method allows us to kill two birds with one stone. First, satisfactory outcomes are shown: the algorithm accuracy reaches 100%, from one hand, and the privacy is protected thanks to the cryptology concept on the other hand.

Utilizing Unlabeled Data to Detect Electricity Fraud in AMI: A Semisupervised Deep Learning Approach.

As nontechnical losses in power systems have recently become a global concern, electricity fraud detection models attracted increasing academic interest. The wide application of smart meters has offered more possibility to detecting fraud from user's consumption patterns. However, the performances of existing consumption-based electricity fraud detection models are still not satisfactory enough for practice, partly due to their limited ability to handle high-dimensional data. In this paper, a deep-learning-based model is developed for detecting electricity fraud in the advanced metering infrastructure, namely, the multitask feature extracting fraud detector (MFEFD). The deep architecture has brought MFEFD a powerful ability to handle high-dimensional input, through which consumption patterns inside load profiles can be effectively extracted. Another challenge is that the insufficiency of labeled data has restricted the generalization of existing models since they are mostly based on supervised learning and labeled data. MFEFD is trained in a semisupervised manner, in which multitask training was implemented to combine the supervised and unsupervised training, so that both the knowledge from unlabeled and labeled data can be made use of. Real-world-data-based case studies have demonstrated MFEFD's high detection performance, robustness, privacy preservation, and practicability.

Evaluating Performance of Wireless Sensor Network in Realistic Smart Grid Environment

Wireless Sensor Networks (WSNs) is a strong candidate for smart grid applications, such as advanced metering infrastructure, demand response management, dynamic pricing, load control, electricity fraud detection, fault diagnostics, substation monitoring and control as well as automation of various elements of the power grid. The realization of these applications directly depends on efficiency of communication facilities among power grid elements. However, the harsh power grid environmental conditions with obstacles, noise, interference, and fading pose great challenges to reliability of these facilities to monitor and control the power grid. The purpose of this paper is to evaluate performance of WSNs in different power grid environments such as 500 kv substations, main power control room, and underground network transformer vaults. The power grid environments are modeled using a log-normal shadowing path loss model channel with realistic parameters. The network is simulated and performance is evaluated using packet delivery ratio, communication delay, and energy consumption. The simulation results have revealed that different environments have considerable impacts on performance of WSNs which make it suitable for most applications that need low data rate with low reliability requirements.

Non-technical losses: detection methods and regulatory aspects overview

Non-technical losses have been reported as one of the most serious problems faced by Electric Utilities. This study provides an overview of most recent methods for electricity fraud detection, based on concepts from data mining, state estimation, game theory and so on. Furthermore, metrics for evaluating such above methods will be discussed and evaluated. It also provides a comparative overview of the main regulatory aspects of frauds based on questionnaire circulated among ten EE countries. Results are discussed and evaluated.

Identification of suspicious electricity customers

In this paper, customer’s invoiced energy time series in the recent past is analyzed in order to detect the cause of increased total electricity losses in the low voltage network. Period from 2003 to 2017 is included in the recent past. This period is organized in five successive decades whose beginnings are shifted for one year. Fuzzy logic is used asa method for determining a set of suspicious electricity customers in the first phase of electricity fraud detecting. At this phase, every customer located in the area of increased total electricity losses is analyzed. For each customer, a time series of invoiced energy are formed. Selected time series data and their relations are used to create fuzzy sets of suspicion. Then, total suspicion value of each customer is determined by using fuzzy logic. Based on the estimated total and technical energy losses in the customer’s area (region that is supplied by one or more MV/LV transformer stations) and the balance of total, invoiced and energy of losses, a boundary value of suspicion percentage is determined. All customers, whose percentage suspicion value is greater than the boundary value, are declared suspect. Thus, suspicious customers with their locations which need inspection are obtained. On-site inspection of suspected customers is not performed and is not the subject of this paper.

Fraud detection in registered electricity time series

Abstract This paper analyses time series without the seasonal component of consumers’ power consumption at low voltage in order to detect fraud and illogical consumption by customers. Statistical process control is used, where the process represents the process of using electricity. XMR charts are used to indicate major changes (decrease) in registered customers’ consumption. Verification of this method was tested on time series of a set of customers who were caught stealing during a time series. It shows that symptoms of non-random factors in time series of customers are revealed in a high percentage, which indirectly confirms the method’s ability to successfully detect electricity fraud.

Using the rough set theory to detect fraud committed by electricity customers

High values of total percentage losses of electricity in Serbia’s distribution network and a very small percentage of success in detecting electricity fraud indicate a need for a sophisticated means of detecting fraud perpetrators. In order for this task to be performed efficiently, the authors of the paper emphasize the need for a suitable and comprehensive use of the billing system or the database of the amounts of invoiced spent electricity in accounting periods and other relevant data regarding registered customers. For that purpose, they suggest using the rough set theory and give a general approach to its use. The point of the paper is forming a criterion for the estimation of accurate (suitable) discretization of original data. The criterion is based on the amount of lost not invoiced electricity due to electricity fraud. Based on consumption characteristics of detected fraud of customers whose measurement points were regularly (monthly) read, a list of the suspicious customers will be formed which will serve as the basis for sending expert teams to specified locations with the task to confirm or dismiss the fraud suspicion.