Fraudulent Consumers Research Articles

Electricity Theft Detection Using Rule-Based Machine Leaning (rML) Approach

Since electricity theft affects non-technical losses (NTLs) in power distribution systems, power companies are genuinely quite concerned about it. Power companies can use the information gathered by Advanced Metering Infrastructure (AMI) to create data-driven, machine learning-based approaches for Electricity Theft Detection (ETD) in order to solve this problem. The majority of data-driven methods for detecting power theft do take usage trends into account while doing their analyses. Even though consumption-based models have been applied extensively to the detection of power theft, it can be difficult to reliably identify theft instances based only on patterns of usage. In this paper, a novel rule-based combined machine learning (rML) technique is developed for power theft detection to address the drawbacks of systems that rely just on consumption patterns. This approach makes use of the load profiles of energy users to establish rules, identify the rule or rules that apply to certain situations, and classify the cases as either legitimate or fraudulent. The UEDAS smart business power consumption dataset's real-world data is used to assess the performance of the suggested technique. Our technique is an innovation in theft detection that combines years of intensive theft tracking with the use of rule-based systems as feature spaces for traditional machine learning models. With an astounding 93% recall rate for the rule-based feature space combination of the random forest classifier, this novel approach has produced outstanding results. The acquired results show a noteworthy accomplishment in the field of fraud detection, successfully detecting fraudulent consumers 77% of the time during on-site examination.

Improving Electricity Theft Detection Using Electricity Information Collection System and Customers’ Consumption Patterns

Electricity theft detection (ETD) techniques employed to identify fraudulent consumers often fail to accurately pinpoint electricity thieves in real time. The patterns associated with electricity use are leveraged to identify anomalies indicative of electricity theft. However, challenges in the benchmark ETD include overfitting and a high incidence of false positives (FPs) resulting from incorrect usage patterns formed by considering only electricity consumption patterns without accounting for external factors that contribute to variations in normal consumption patterns. Further investigation is required to precisely detect instances of electricity theft, thereby minimizing nontechnical losses and forecasting future electricity demand to maintain a stable supply. This study employs a master energy meter located on the transformer side to monitor the amount of energy distributed to the region. Zones with a high likelihood of energy theft are detected by comparing the sum of readings from all the smart meters with the readings from the master energy meter installed on the HV of the substation transformer within the same electric feeder. Ensemble XGBoost machine-learning algorithm and K-Means algorithm are used for the classification of malicious and nonmalicious samples and grouping similar types of consumers together, respectively. This makes the proposed model resistant to false-positive rates caused by changes in usage patterns that aren’t done on purpose. Furthermore, energy thieves are identified by detecting anomalies in consumption behavior while maintaining constant internal and external environmental variables. This novel model proposed here mitigates the FP rate found in the present research of electricity usage data. Our approach outperforms support vector machines, convolution neural network, and logistic regression in simulations. Precision, F1-score, recall, Matthews Correlation Coefficient, Receiver Operating Characteristics (ROC)-Area Under The Curve (AUC), and Precision Recal (PR)-Area Under The Curve (AUC) validate our model. The simulation results show that the proposed K-Means-LSTM-XGBoost model improved the classifier’s F1-score, precision, and recall to 93.75%, 95.16%, and 92.38%, respectively. Our model classifies huge time series data with high precision and can be utilized by the utility for real time electricity theft detection.

Detection of Non-Technical Losses in Irrigant Consumers through Artificial Intelligence: A Pilot Study

Non-technical losses (NTLs) verified in the power distribution grids cause great financial losses to power utilities. In rural distribution grids, fraudulent consumers contribute to technical problems. The Southern region in Brazil contains more than 70% of the total rice production and power irrigation systems. These systems operate seasonally in distribution grids with high NTL conditions. This work aimed to present an artificial intelligence-based system to help power distribution companies detect potential consumers causing NTLs. This minimizes the challenge of maintaining compliance with current regulations and ensuring the quality of services and products. In the proposed methodology, historical energy consumption information, meteorological data, satellite images, and data from energy suppliers are processed by artificial intelligence, indicating the suspicious consumer units of NTL. This work presents every step developed in the proposed methodology and the tool application in a pilot area. We detected a high number of consumers responsible for NTLs, with an accuracy of 63% and an average reduction of 78% in the search area. These results corroborated the effectiveness of the tool and instigated the research team to expand the application to other rice production areas.

Machine intelligence aware electricity theft detection for smart metering applications

Electricity theft detection (ETD) is a serious issue and needs proper attention to investigate and detect the fraudulent consumers. Fraudulent consumers steal electricity and burdensome utility providers, which causes huge revenue losses. To minimize such revenue loss an affine investigation and detection scheme is required. In this paper, we use AttenLSTMInception in order to investigate the electricity theft scenario. Initially, benign consumers data is manipulated through false data injection techniques (FDIs). Six various types of FDIs are used to manipulate the benign class data, which synthesizes six variants of the manipulated data. The benign and synthesized variants are concatenated and the data is fed to borderline SMOTESVM. In order to tackle the issue of synthesizing data from overlapped samples, borderline SMOTESVM is used. The balanced data is segmented into training and testing data. Furthermore, Matthews correlation coefficient is used to investigate model's performance by measuring confidence score of the binary classification. Moreover, to evaluate the robustness and to validate the performance and of the model, testing data of 20%, 40% and unseen is used. Similarly, six activation functions are used to evaluate the model's robustness. Furthermore, impact of the imbalanced data is studied using MCC score. Support vector machine (SVM), random forest (RF), decision tree (DT), convolutional neural network-based long short-term memorization (CNN-LSTM) are used as base models. Our model outperforms the base models by achieving high recall, precision, accuracy, AUC-score and F1 score. These values for our proposed model are 0.94%, 0.94%, 0.98%, 0.98%, respectively.

A novel technique for detecting electricity theft in secure smart grids using CNN and XG-boost

Electricity theft is one of the main nontechnical losses (NTLs) in distributed networks which cause significant harm to the power grids. As power grids provide the centralized power to all the connected consumers, therefore, any fraudulent consumption can cause harm to the power grids which can damage the whole electric power supply and can influence its quality. The detection of such fraudulent consumers becomes difficult when there is a large amount of data. Smart grids can be used to solve this problem as it provides a two-way electricity flow which allows someone to detect, reenact and apply new changes to the electric data flow. The existing systems for electricity theft detection, works on the principle of one dimensional (1-D) electric data, which provides poor accuracy in theft detection. Therefore, an ensemble model based on convolutional neural network and extreme gradient boosting (CNN-XGB) model is presented in this paper. In this model both one dimensional (1-D) and two-dimensional (2-D) electricity consumption data are used to pass to the CNN model. Proposed model achieved the accuracy of 92% for electricity theft detection, which is better than existing models.

Towards Efficient Energy Utilization Using Big Data Analytics in Smart Cities for Electricity Theft Detection

In energy sectors, power utilities face financial losses due to Electricity Theft (ET). It happens when electricity is consumed without billing. Several methods are developed to detect ET automatically. Most of these methods only assess Electricity Consumption (EC) records. However, it is challenging to detect fraudulent consumers by only observing EC records because of diverse theft strategies (line tapping, meter tampering, etc.) and the irregularity of ET behavior. Furthermore, many methods have poor classification accuracy due to imbalanced data. This work proposes two novel methods to resolve the above-mentioned issues: Tomek Link Borderline Synthetic Minority Oversampling Technique with Support Vector Machine (TBSSVM) and Temporal Convolutional Network with Enhanced Multi-Layer Perceptron (TCN-EMLP). The former resamples the data by balancing the majority and minority class instances. Whereas, the latter classifies normal and fraudulent consumers. Moreover, deep learning models suffer from high variance in their final results due to the assignment of different weights. Therefore, an averaging ensemble strategy is applied in this work to reduce the high variance. Furthermore, State Grid Cooperation of China (SGCC) and Pakistan Residential Electricity Consumption (PRECON) datasets are used in this paper for performing the simulations. SGCC is an imbalanced and labeled dataset while PRECON is an unlabeled dataset comprised of normal consumers' EC records (sequential) and auxiliary (non-sequential) data. Simulation results show that the proposed model outperforms the baselines, i.e., wide and deep convolutional neural network, extreme gradient boosting, long short-term memory with multi-layer perceptron, etc., in terms of ET detection.

Real-Time Detection of False Readings in Smart Grid AMI Using Deep and Ensemble Learning

In the advanced metering infrastructure, smart meters are deployed at the consumers’ side to regularly transmit fine-grained electricity consumption readings to the system operator (SO) for billing and real-time load monitoring and energy management. However, fraudulent consumers may compromise their meters to launch electricity-theft cyberattacks by reporting low-consumption readings to reduce their bills. These false readings not only cause financial losses but also degrade the grid’s performance because they are used for energy management and load estimate. The existing solutions in the literature focus only on securing the billing, so they are not designed to detect the attacks in real time, and thus the SO may use false readings for a long period of time in load monitoring and energy management until they are identified. In this paper, we propose a general ensemble-based deep-learning detector that enables the SO to detect false readings in real time. To do that, we first train several deep learning models on samples generated from a sliding window of the readings. Then, we use the best-performing model to train several models on different ratios of false readings and use them in our ensemble-based detector. Extensive experiments are conducted, and the results indicate that comparing to the literature, our detector can detect the false readings after sending a few false readings (around 15) comparing to the existing daily and weekly detection approaches that need 144 and 1,008 readings, respectively.

An adaptive synthesis to handle imbalanced big data with deep siamese network for electricity theft detection in smart grids

The bi-directional flow of energy and information in the smart grid makes it possible to record and analyze the electricity consumption profiles of consumers. Because of the increasing rate of inflation over the past few years, people started looking for means to use electricity illegally, termed as electricity theft. Many data analytics techniques are proposed in the literature for electricity theft detection (ETD). These techniques help in the detection of suspected illegal consumers. However, the existing approaches have a low ETD rate either due to improper handling of the imbalanced class problem in a dataset or the selection of inappropriate classifier. In this paper, a robust big data analytics technique is proposed to resolve the aforementioned concerns. Firstly, adaptive synthesis (ADASYN) is applied to handle the imbalanced class problem of data. Secondly convolutional neural network (CNN) and long-short term memory (LSTM) integrated deep siamese network (DSN) are proposed to discriminate the features of both honest and fraudulent consumers. Specifically, the task of feature extraction from weekly energy consumption profiles is handed over to the CNN module while the LSTM module performs the sequence learning. Finally, the DSN contemplates on the shared features provided by the CNN-LSTM and applies final judgment. The data analytics is performed on different train–test ratios of the real-time smart meters’ data. The simulation results validate the proposed model’s effectiveness in terms of high area under the curve, F1-Score, precision and recall.

A Novel Features-Based Multivariate Gaussian Distribution Method for the Fraudulent Consumers Detection in the Power Utilities of Developing Countries

According to statistics, developing countries all over the world have suffered significant non-technical losses (NTLs) both in natural gas and electricity distribution. NTLs are thought of as energy that is consumed but not billed e.g., theft, meter tampering, meter reversing, etc. The adaptation of smart metering technology has enabled much of the developed world to significantly reduce their NTLs. Also, the recent advancements in machine learning and data analytics have enabled a further reduction in these losses. However, these solutions are not directly applicable to developing countries because of their infrastructure and manual data collection. This paper proposes a tailored solution based on machine learning to mitigate NTLs in developing countries. The proposed method is based on a multivariate Gaussian distribution framework to identify fraudulent consumers. It integrates novel features like social class stratification and the weather profile of an area. Thus, achieving a significant improvement in fraudulent consumer detection. This study has been done on a real dataset of consumers provided by the local power distribution companies that have been cross-validated by onsite inspection. The obtained results successfully identify fraudulent consumers with a maximum success rate of 75%.

Managing consumer returns with technology-enabled countermeasures

This paper examines retail return abuse with respect to both opportunistic and fraudulent consumer behavior. The decisions of interest are the retailer’s price for purchases and refund amount for returns. Our analysis provides managerial insight into how a retailer makes these decisions to mitigate return abuse. Including both forms of return abuse in a base model, we find that there is an interaction effect that meaningfully changes a retailer’s optimal decisions and profit, from what would be observed if only opportunism was present. We also evaluate two innovative technology-enabled countermeasures designed to mitigate return abuse: customer profiling and product tracking. A customer profiling system identifies opportunistic customers by using their personal identification and transaction history. In contrast, a product tracking system identifies fraudulent returns by recording each transaction of a product through the use of unique identifiers. We develop prescriptive models for these technologies and investigate the value of making such investments. Our analyses demonstrate the conditions in which it is advantageous to adopt these technologies and when such investments should be avoided. We find that when a retailer is able to charge restocking fees, it is well equipped to mitigate opportunism and fraud with only its price and refund decisions. Hence, the value of consumer profiling and product tracking technologies is limited. However, when a retailer is constrained to offer a full refund due to market dynamics, employing these technologies may hold significant value. Our analysis details the determinants of value, model sensitivity, and comparisons between models. We also address how these countermeasures impact a retailer’s profitability, demand structure, and decisions with respect to price and refund amount.

Efficient Privacy-Preserving Electricity Theft Detection With Dynamic Billing and Load Monitoring for AMI Networks

In advanced metering infrastructure (AMI), smart meters (SMs) are installed at the consumer side to send fine-grained power consumption readings periodically to the system operator (SO) for load monitoring, energy management, and billing. However, fraudulent consumers launch electricity theft cyber attacks by reporting false readings to reduce their bills illegally. These attacks do not only cause financial losses but may also degrade the grid performance because the readings are used for grid management. To identify these attackers, the existing schemes employ machine-learning models using the consumers' fine-grained readings, which violates the consumers' privacy by revealing their lifestyle. In this article, we propose an efficient scheme that enables the SO to detect electricity theft, compute bills, and monitor load while preserving the consumers' privacy. The idea is that SMs encrypt their readings using functional encryption (FE), and the SO uses the ciphertexts to: 1) compute the bills following the dynamic pricing approach; 2) monitor the grid load; and 3) evaluate a machine-learning model to detect fraudulent consumers, without being able to learn the individual readings to preserve consumers' privacy. We adapted an FE scheme so that the encrypted readings are aggregated for billing and load monitoring and only the aggregated value is revealed to the SO. Also, we exploited the inner-product operations on encrypted readings to evaluate a machine-learning model to detect fraudulent consumers. The real data set is used to evaluate our scheme, and our evaluations indicate that our scheme is secure and can detect fraudulent consumers accurately with low communication and computation overhead.

Monitoring technical losses to improve non-technical losses estimation and detection in LV distribution systems

Abstract The process of mitigating non-technical losses (NTL) in power distribution utilities is done in two stages. The first determines which distribution transformers have high NTL values. The second attempts to locate fraudulent consumers, powered by these transformers. This paper proposes a new methodology to improve the calculation of technical losses (TL), leading to a better estimation of the NTL, using temperature sensors. It is also presented a new process to identify possible energy theft locations using voltage drop differences. The identification of possible energy pilfering spots is done with the aid of the measurement of technical losses obtained in the first stretches of the low voltage network, near the transformer, where the TL is more significant. A Backward/Forward Sweep algorithm using the power summation technique is used to calculate the voltage drops in two situations: with power data read only from the smart meters and power data including the TL readings. An analysis of the differences in voltage drop at each point between the two situations makes it possible to locate the probable energy thief.

Fraud Detection for Metered Costumers in Power Distribution Companies Using C5.0 Decision Tree Algorithm

Non-technical Losses mainly Electricity theft has been a main concern for power utilities from last many years. Power utilities are estimated to lose billion dollars annually because of illegal usage of electricity by fraudulent consumers. Researchers are trying different methods for proficiently recognizing fraudster costumers. This research suggests a new approach based on C5 algorithm for efficiently identifying consumers involved in electricity theft. The C5.0 algorithm is a modified form of the C4.5 algorithm. It is also one of the decision tree algorithms but with a much-improved classification rate. The C5.0 algorithm relies on monthly energy consumption data to identify any anomaly in consumer energy usage data associated with NTL behavior. There are many types of fraud committed by fraudulent consumers but this research is focused on fraudulent consumers who have a unexpected deviation from their usual load profile. The motivation of this research is to aid Power distribution companies in Pakistan to decrease there NTL’s due to pilfering in energy consumption by fraudulent consumers. The accuracy of the C5.0 algorithm is 94.61% which is much higher when compared to some state of the art machine learning algorithms like Random forest, Support Vector Machine, K-NN and other decision trees.

Non-technical Losses Detection: An Innovative No-Neutral Detector Device for Tampered Meters

The increase in electric energy tariffs in conjunction with social and cultural aspects often leads electric energy consumers to tamper with electric energy meters, as a means to decrease the reading for consumed energy at their respective residencies. One form of meter tampering that is very difficult to detect is the removal of the neutral conductor, which goes on to produce an effect where the meter registers a consumption that is lower than the real consumption. The identification of this fraud is difficult due to the fact that the readings are scheduled, and as such the fraudulent consumer needs only to reconnect the neutral conductor at the time of the meter reading. This article presents a device developed for identifying and registering this type of theft, thus making it more difficult to practice. Laboratory tests and simulations have shown that the lower the power factor of an installation in low voltage, lower will be the reading registered on the meter when this type of fraud arises. This newly developed device has shown itself to be robust, with a low cost as well as efficient in performing to its proposed intent.

The management of first party fraud in e-tailing: a qualitative study

Purpose First-party fraud in which retail consumers commit fraud against retailers is a growing problem. However, to date studies on retail crime have focused almost entirely on fraudulent consumer behaviours in physical stores. With the growth of e-commerce, the losses to retailers from this fraud are growing so there is strong need to research this problem from multiple perspectives. The paper aims to discuss this issue. Design/methodology/approach The authors conducted three case studies and a total of 24 semi-structured interviews with retail managers, and evaluated their existing prevention-related documentation. Fraud management lifecycle theory was used to organise and discuss the findings. Findings The authors found that many retailers are treating this problem as just a cost of doing business online and have no detailed plans for dealing with this problem or any reporting to law enforcement agencies. However, they have begun working with delivery companies for delivery accuracy. Use of convenience stores as collection points is also showing early improvements. Research limitations/implications The small number of cases and interviews used is a limitation of this study. However, the authors believe that the findings are useful for advancing knowledge in this emerging research area. Practical implications This study provides insight into existing management practices in this domain, and makes recommendations on how to improve the management of first-party fraud. The study also makes a case for increased managerial interest and involvement in reducing first-party fraud. The study also helps bridge a glaring gap in existing literature and provides useful leads for further research. Originality/value To the authors’ knowledge, this is the first study to evaluate the existing practices employed to manage first-party fraud in e-retail.

PPETD: Privacy-Preserving Electricity Theft Detection Scheme With Load Monitoring and Billing for AMI Networks

In advanced metering infrastructure (AMI) networks, smart meters installed at the consumer side should report fine-grained power consumption readings (every few minutes) to the system operator for billing, real-time load monitoring, and energy management. On the other hand, the AMI networks are vulnerable to cyber-attacks where malicious consumers report false (low) electricity consumption to reduce their bills in an illegal way. Therefore, it is imperative to develop schemes to accurately identify the consumers that steal electricity by reporting false electricity usage. Most of the existing schemes rely on machine learning for electricity theft detection using the consumers’ fine-grained power consumption meter readings. However, this fine-grained data that is used for electricity theft detection, load monitoring, and billing can also be misused to infer sensitive information regarding the consumers such as whether they are on travel, the appliances they use, and so on. In this paper, we propose an efficient and privacy-preserving electricity theft detection scheme for the AMI network and we refer to it as PPETD. Our scheme allows system operators to identify the electricity thefts, monitor the loads, and compute electricity bills efficiently using masked fine-grained meter readings without violating the consumers’ privacy. The PPETD uses secret sharing to allow the consumers to send masked readings to the system operator such that these readings can be aggregated for the purpose of monitoring and billing. In addition, secure two-party protocols using arithmetic and binary circuits are executed by the system operator and each consumer to evaluate a generalized convolutional-neural network model on the reported masked fine-grained power consumption readings for the purpose of electricity theft detection. An extensive analysis of real datasets is performed to evaluate the security and the performance of the PPETD. Our results confirm that our scheme is accurate in detecting fraudulent consumers with privacy preservation and acceptable communication and computation overhead.

NTL Detection in Electric Distribution Systems Using the Maximal Overlap Discrete Wavelet-Packet Transform and Random Undersampling Boosting

The illegal use of electricity, defective meters, and a malfunctioning infrastructure are major causes of Non-technical losses (NTLs) in electric distribution systems. Although the use of supervised machine learning techniques to detect NTLs has been widely studied, further research is needed in order to address some significant challenges. (i) Given that fraudulent consumers remarkably outnumber non-fraudulent ones, the imbalanced nature of the dataset can have a major negative impact on the performance of supervised machine learning methods. (ii) Given the large number of dimensions present in the time series data used for training and testing classifiers, advanced signal processing techniques are required in order to extract the most relevant information. (iii) The effectiveness of classifiers must be evaluated using meaningful performance measures for imbalanced data . This paper proposes a framework that addresses the three previous challenges. The core of the proposed framework is the application of the maximal overlap discrete wavelet-packet transform (MODWPT) for feature extraction from time series data and the random undersampling boosting (RUSBoost) algorithm for NTL detection. Moreover, our framework is evaluated using an extensive list of performance metrics. Experiments show that the MODWPT combined with the RUSBoost algorithm can significantly improve the quality of NTL predictions.

An anomaly detection framework for identifying energy theft and defective meters in smart grids

Smart meters are progressively deployed to replace its antiquated predecessor to measure and monitor consumers’ consumption in smart grids. Although smart meters are equipped with encrypted communication and tamper-detection features, they are likely to be exposed to multiple cyber attacks. These meters may be easily compromised to falsify meter readings, which increases the chances and diversifies the types of energy theft. To thwart energy fraud from smart meters, utility providers are identifying anomalous consumption patterns reported to operation centers by leveraging on consumers’ consumption data collected from advanced metering infrastructure. In this paper, we put forward a new anomaly detection framework to evaluate consumers’ energy utilization behavior for identifying the localities of potential energy frauds and faulty meters. Metrics known as the loss factor and error term are introduced to estimate the amount of technical losses and capture the measurement noise, respectively in the distribution lines and transformers. The anomaly detection framework is then enhanced to detect consumers’ malfeasance and faulty meters even when there are intermittent cheating and faulty equipment, improving its robustness. Results from both simulations and test rig show that the proposed framework can successfully locate fraudulent consumers and discover faulty smart meters.

Detection of energy theft and defective smart meters in smart grids using linear regression

The utility providers are estimated to lose billions of dollars annually due to energy theft. Although the implementation of smart grids offers technical and social advantages, the smart meters deployed in smart grids are susceptible to more attacks and network intrusions by energy thieves as compared to conventional mechanical meters. To mitigate non-technical losses due to electricity thefts and inaccurate smart meters readings, utility providers are leveraging on the energy consumption data collected from the advanced metering infrastructure implemented in smart grids to identify possible defective smart meters and abnormal consumers’ consumption patterns. In this paper, we design two linear regression-based algorithms to study consumers’ energy utilization behavior and evaluate their anomaly coefficients so as to combat energy theft caused by meter tampering and detect defective smart meters. Categorical variables and detection coefficients are also introduced in the model to identify the periods and locations of energy frauds as well as faulty smart meters. Simulations are conducted and the results show that the proposed algorithms can successfully detect all the fraudulent consumers and discover faulty smart meters in a neighborhood area network.

First party fraud: a review of the forms and motives of fraudulent consumer behaviours in e-tailing

Purpose – First party fraud is fraud committed by an account holder or customer that does not involve the use of a stolen identity. This type of fraud has grown substantially in recent times due to increased online shopping and is becoming a major concern for online retail businesses, hereby referred to as e-tailers. The purpose of this paper is to provide insights into the motives and nature of first party fraud in e-tailing. Design/methodology/approach – Systematic literature review was used to synthesise existing research on first party fraud. The authors used scholarly literature as well as grey literature to help understand the motives and nature of this growing business problem. Findings – Findings reveal a myriad of schemes and motives for engaging in first party fraud. Research limitations/implications – The findings of this paper employed very little literature due to availability. However, the authors believe that the findings are still useful for advancing the knowledge in this emerging research area. Practical implications – This study will be useful to researchers as well as practitioners in the retail industry in helping understand the nature and motives of first party frauds which could in turn help devise preventive strategies. The study also makes a case for increased managerial interest and involvement in reducing first party fraud. Originality/value – A comprehensive literature search presented in this paper shows that this is the first paper to synthesise the various forms of first party fraud in e-tailing.