Electricity Theft Research Articles

Data-oriented ensemble predictor based on time series classifiers for fraud detection

Non-Technical Losses (NTL) remain a challenge because electricity theft is prevalent in traditional meters and smart metering systems and structures. They cause financial losses that affect the security of supply and lead to a cooperative burden because NTL are incorporated in the expense tariff and are paid by all customers in countries such as India, China, Brazil, Tunisia, Uruguay, and others. A significant number of methods have been proposed to detect NTL, but they cannot precisely recognize this type of fraud, which enables companies to better identify the fraud’s origin and evaluate its financial impact. In addition, it is important to combine several Time-Series Classification (TSC) algorithms which consider the time-dependent nature of energy consumption data. We propose a data-oriented heterogeneous ensemble predictor based on time series classifiers for NTL detection, called DETECT. It classifies energy consumption samples either as honest or as fraudulent and further categorizes the type of fraud which might have been committed based on the analysis of the consumption patterns of each sample. DETECT is a fraud detection algorithm that has the ability to analyze and classify NTL frauds effectively regardless of the dataset. Using our proposed approach, we obtained a performance improvement with a Detection Rate (DR) value equal to 93.45% and a False Positive Rate (FPR) value equal to 1.61%. DETECT focuses on time-series data, which enables the development of a method that better interprets real-world scenarios and it is more error-resistant.

A Review on MPC Based Self Recovering Intelligent Advance Meter for Smart Grid: Scheme and Challenges

The Model Predict Control (MPC) based Intelligent Advance Metering (IAM) is a core maneuver of future smart grids (SG). SG is the advanced generation of electric power and utility system that improve operation technology (OT) and information technology (IT) to provide nonstop, self-recovery, self-configuration, low-cost, and security-based electricity to the consumer in real-time. Smart metering (SM) allows SG to connect the electric, gas, and oil utilities through sensors. Power plants, consumers, and utility companies will be received real-time wireless control IAM with fifth generation (5G) network technology. The aim of 5G network technology is to enable power grid digitalization (PGD) and facilitate the (IOT) Internet of Things for the future advance SG with benefits such as high-rate public safety, low latency, ultra-high speed, large number of connectivity, and reliability. In this paper, we analyze future predictions about energy needs by using MPC, fast self-recovery system, self-configuration, and upgradation, better performance of service provider, faster power connecting after an outage, control electric theft, minimize electric leakage, a large number of wireless connecting of IAM home-based, and real-time monitoring via human machine interface (HMI) and for customer end IAM operation over 5G networks to reduce billing price, reduce meter cost, lower outage cost, and as well as personalized control over electricity consumption and future challenge in this area.

"CAPABILITIES OF NB-IoT TECHNOLOGY FOR SOLVING TASKS OF MONITORING, ACCOUNTING AND COLLECTING DATA ON ELECTRICITY CONSUMPTION "

"Problems with accounting, collection and correct readings of electricity meters are of great importance when preparing and invoicing for electricity, especially in neighborhoods with private households, where the process of monitoring electricity consumption is not always provided for various reasons beyond the control of the consumer of services. The development of modern information technologies, including mobile communications using the technology of ΝΒ-ΙοΤ, allows us to solve problems related to improving the efficiency of monitoring, accounting, and data collection of electricity consumption in microdistricts not only quickly and efficiently, but also increase the management efficiency of the process of collecting meter readings, as well as reducing the costs of the Service Provider for transport services, personnel and other items of expenditure. The article presents an analysis of various brands of metering devices used in the Urker microdistrict and the use of metering devices with the support of technology ΝΒ-ΙοΤ. A simulation model of access to metering devices using the technology of ΝΒ-ΙοΤ is shown. The issues related to the causes and problems of electricity theft are touched upon. The article will be useful to students, doctoral students, engineers, anyone who is related to automation and control of technological processes and productions, as well as anyone who deals with the problems of implementing the ASUE, IoT technology. "

Employing Feature Extraction, Feature Selection, and Machine Learning to Classify Electricity Consumption as Normal or Electricity Theft

Employing Feature Extraction, Feature Selection, and Machine Learning to Classify Electricity Consumption as Normal or Electricity Theft

A Convolution–Non-Convolution Parallel Deep Network for Electricity Theft Detection

This paper proposes a novel convolution–non-convolution parallel deep network (CNCP)-based method for electricity theft detection. First, the load time series of normal residents and electricity thieves were analyzed and it was found that, compared with the load time series of electricity thieves, the normal residents’ load time series present more obvious periodicity in different time scales, e.g., weeks and years; second, the load times series were converted into 2D images according to the periodicity, and then electricity theft detection was considered as an image classification issue; third, a novel CNCP-based method was proposed in which two heterogeneous deep neural networks were used to capture the features of the load time series in different time scales, and the outputs were fused to obtain the detection result. Extensive experiments show that, compared with some state-of-the-art methods, the proposed method can greatly improve the performance of electricity theft detection.

Economic Research of Electricity Losses in PLN Due to Theft Differences with Methods Changing MCB and Influencing Measurement in kWh Meter

Today, there were many cases of theft or infringement of electricity in various ways. This was caused PLN to suffer losses due to the theft of electricity amounting to Rp 1.5 trillion per year. Electrical theft focused on kWh meter there were various kinds, such as by changing the side of the MCB and adding a needle on the kWh meter. The objective of this study were found out the recorded kWh ratio if the installation was installed normally with a modified installation by changing on the MCB side, adding a needle, and comparing the economic value that PLN should had obtained when the kWh meter was installed normally and the kWh meter was changed. The method used in the research for normal measurement as much as 6 times the rotation of the disk and used 5 variations of load ; the measurement by changing the MCB and adding the needle was the same as the normal measurement, only when changing the MCB changes 2 A to 4 A, and when adding the needle, the disk was plugged with a needle. The result showed that the average of error difference between kWh meter that was installed normally with that has been changed on the side of MCB is 0, and kWh meter installed with normal that has been changed add needle equal to -44.1586. The difference in total cost of electricity for one month for normal kWh meter with which has been changed on the MCB side was Rp. 165.914.-, and for the normal kWh meter difference with the added needle of Rp. 58.610.

A Novel Evasion Attack Against Global Electricity Theft Detectors and a Countermeasure

The smart grid advanced metering infrastructure (AMI) is vulnerable to electricity theft cyber-attacks in which malicious smart meters report low readings to reduce the consumers’ bills. To avoid this problem, several machine-learningbased detectors have been proposed to detect electricity theft. Most of these detectors are global in the sense that they are trained on different consumption levels, including low and high consumptions, to be used for all consumers. In this paper, we introduce a novel type of evasion attacks against global detectors as follows. A malicious consumer who has high consumption level can send false readings for a low consumption profile (that resembles the profiles the detector is trained on) to evade the detector, i.e., steal electricity without being detected. We firstly conduct experiments to prove that the existing global detectors are vulnerable to this new kind of evasion attacks. To launch this attack, we train a Generative Adversarial Network (GAN) on a real dataset to generate fake low consumption readings that can evade the detector. The given results indicate that the success rate of the attack is between 82% and 97%. To thwart this attack, we divide the consumers into clusters of close electricity consumption levels and train one detector for each cluster. Therefore, if a malicious consumer in any cluster tries to imitate the consumption profiles of consumers in other clusters, he/she will be detected. On the other hand, it is not profitable to imitate the electricity consumption profiles of consumers in his/her cluster to evade detection. To prove the effectiveness of our countermeasure, extensive experiments are conducted and the results indicate that our countermeasure can successfully thwart the attack.

Electricity theft detection based on hybrid random forest and weighted support vector data description

Improving the detection rate of electricity theft users in smart grids is crucial to the safe operation of the power system and the economic efficiency of the grid. The traditional electricity theft detection relies on manual household inspection, which is costly and inefficient. With the application of smart meters, machine learning-based methods are gradually applied to detect electricity theft users. However, the training of the model is limited by the unbalanced electricity consumption dataset, and the performance of the model is limited in the process of processing high-dimensional data. Therefore, this paper adopts ensemble learning and imbalance data classification model to analyze the electricity theft detection problem in complex grid environment, and proposes an electricity theft detection model based on random forest and weighted support vector description. Firstly, the feature selection algorithm for mixed load data with statistical features based on random forest (FS-MLDRF) is used to extract the features that can best represent the customers’ electricity consumption behavior, so as to reduce the complexity of user data. Finally, an electricity theft detection based on hybrid random forest and weighted support vector data description is constructed (HFR-WSVDD), and the SVDD classifier is trained with the features weighted by feature importance to improve the performance of power theft detection in the case of unbalanced data. The simulation experiment results on the real electricity dataset show that compared with existing feature selection algorithms and anomaly detection algorithms, the algorithms proposed in this paper have excellent feature representation capability and better classification performance.

Outage-induced power backup choice in Pakistan

Like in many other developing countries, power outages are prevalent in Pakistan. Consumers are compelled to explore alternative energy strategies when the overhead of the outage surpasses that of implementing a backup plan. Inverters, small generators, and solar energy systems (SESs) are among the common backup options. This study utilizes primary data collected from the twin cities of Islamabad and Rawalpindi to delve into the factors influencing consumers' perception-based choices regarding backup solutions. To account for various backup choices, we employ multinomial logistic regression techniques on a sample of 952 households. Our analysis reveals that perception-based variables, such as electricity theft and decreasing satisfaction, negatively impact the likelihood of selecting off-grid renewable options like solar technology. Moreover, we observe that consumers residing in rented houses prefer inverters and generators, while homeownership significantly influences the adoption of all three backup choices. Interestingly, the rural-urban location of the household does not influence the decision to opt for a SES. These findings carry important policy implications for energy-deficient regions. The transition towards solar energy is driven by the household's strategy to ensure uninterrupted energy backup while simultaneously reducing the cost of electricity consumption.

Real-time power theft monitoring and detection system with double connected data capture system

Power utilities worldwide are facing enormous challenges when it comes to the distribution of electricity. With these challenges, electricity theft is regarded as the most common challenge in the electrical distribution system. Electricity theft can be meter tampering done in consumer houses and illegal connections done using hook-ups from the distribution pole grids. These electricity theft challenges have caused power utilities to reconsider customer engagements focusing on feedback, putting loss detection systems in their distribution system networks, using artificial intelligence to schedule maintenance and other asset management activities, etc. The main focus of this paper is to design a real-time power theft monitoring and detection system that is able to detect power theft in distribution systems. This proposed system utilizes smart meters consisting of an Arduino ATMega328P microcontrollers with GSM modules (Global System for Mobile Communication) used for system communication. Cloud storage is created to store the smart meter data. Simulations of the proposed system were done using Proteus Design Suite v.8.10 SP3 software. The proposed system is practically constructed for prototype measurement results. Should power imbalances be measured by the system, the authority office will receive an SMS notification as an alert for power theft detected by a specific smart metering system. The authority office will analyse the power measurements sent to the cloud storage (MATLAB Online, ThingSpeak IoT channels display), and further action will be taken.

Social Learning Process in Household Electricity Theft: Case Study in Tanjung Burung Village, Teluknaga District, Tangerang Regency

This study analyzes the collective theft of electricity in Tanjung Burung Village. This study uses descriptive qualitative research methods with data collection methods in the form of observation, interviews and documentation. The purpose of this study was to determine the processes and factors behind the collective occurrence of electricity theft. The theory used is social learning theory from Ronald Akers which focuses on four main concepts, namely differential association, definition, differential reinforcement, and imitation. In the theory put forward by Ronald Akers, there are values ​​of crime that are transmitted due to contact access that supports criminal behavior. The results of this study indicate that electricity theft in Tanjung Burung Village occurs due to differential associations where a person is confronted with profitable behavior to commit electricity theft so that electricity theft is considered a justifiable definition and not too bad to do. In addition, in electricity theft there is a differential amplifier in the form of rewards and punishments obtained. The punishments and rewards that are obtained by the perpetrators are generally not a deterrent because there are various ways to take peaceful paths so that the perpetrators can still be electrified illegally. The theft of electricity that continues to occur is motivated by the imitative behavior of previous perpetrators and is supported by weak laws or laws that don't work as they should, so that electricity theft is carried out collectively. This is done because the profits obtained are more dominant than the losses.

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.

Smart Grid Theft Detection Based on Hybrid Multi-Time Scale Neural Network

Despite the widespread use of artificial intelligence-based methods in detecting electricity theft by smart grid customers, current methods suffer from two main flaws: a limited amount of data on electricity theft customers compared to that on normal customers and an imbalanced dataset that can significantly affect the accuracy of the detection method. Additionally, most existing methods for detecting electricity theft rely solely on one-dimensional electricity consumption data, which fails to capture the periodicity of consumption and overlooks the temporal correlation of customers’ electricity consumption based on their weekly, monthly, or other time scales. To address the mentioned issues, this paper proposes a novel approach that first employed a time series generative adversarial network to balance the dataset by generating synthetic data for electricity theft customers. Then, a hybrid multi-time-scale neural network-based model was utilized to extract customers’ features and a CatBoost classifier was applied to achieve classification. Experiments were conducted on a real-world smart meter dataset obtained from the State Grid Corporation of China. The results demonstrated that the proposed method could detect electricity theft by customers with a precision rate of 96.64%, a recall rate of 96.87%, and a significantly reduced false detection rate of 3.77%.

P2Detect: Electricity Theft Detection With Privacy Preservation for Both Data and Model in Smart Grid

Electricity Theft Detection (ETD) based on deep learning can detect abnormal electricity consumption behaviors by analyzing user historical data. However, existing ETD schemes conduct neural network inferencing in plaintext without strong privacy guarantees, moreover, none has considered the privacy of proprietary models from electric utilities. To address those issues, we introduce p2Detect, a privacy-preserving ETD scheme via neural network inferencing over encrypted data. p2Detect works in a two-party setting where model is held by Detection Server (DS) and data are held by Edge Server (ES) on behalf of multiple Smart Meters (SM). First, we design a homomorphic-encryption-friendly model and its corresponding training methods based on region search and clustering; Second, our customized cryptographic protocols enable privacy-preserving yet efficient ETD inferencing over encrypted data, without revealing private inputs to DS or revealing model parameters to ES. The performance evaluations demonstrate: (i) Model optimizations achieve AUC as high as 0.763, where AUC is a robust evaluation criterion for inferencing ability of the model on unbalanced dataset. (ii) p2Detect takes 73.3ms to execute ETD per SM on average. (iii) Communication cost between DS and ES is 196.76KB per SM. Therefore, p2Detect preserves strong privacy while maintaining high accuracy and efficiency.

IOT BASED REDUCTION OF ELECTRICITY THEFT

The Internet of Things (IoT) has emerged as a promising technology to address various challenges in different domains. One of the critical challenges in the power sector is the issue of electricity theft, which results in huge revenue losses for power utilities. To avoid such issues, the proposed system monitors the real-time electricity consumption of a household or commercial building using a current sensor and sends the data to a cloud- based IoT platform, ThingSpeak, using NodeMCU and WiFi. The platform analyzes the data and identifies any unusual electricity consumption patterns, which could be indicative of electricity theft. Keywords- IOT (Internet of Things), Electricity, EnergyMeter Coil, ESP8266, ThingSpeak.

Electricity Theft Detection in Power Consumption using Superiority of Machine Learning Algorithm

Electricity theft is the act of stealing electricity by deceptive techniques. Electricity theft represents a large chunk of NTL (Non- Technical Loss). These are the losses caused by misidentified, mis-allocated, or incorrect energy flows. Electricity theft is a major issue in India, as it is in most of the developing countries. Although the theft can be detected using machine learning techniques. In this study, the system is proposed that reads smart meter data (input) using OCR (Optimized Character Recognition). OCR is a technique for recognizing text characters in digital images that have been printed or handwritten. The smart meter data image which consists of electricity usage units is converted into machine- readable text by OCR. Furthermore, the SARIMAX (Seasonal Auto Regressive Integrated Moving Average with exogenous factors) algorithm is utilized to monitor customers electricity consumption and detect electricity theft. If theft is identified, an alert message and details of the theft are sent to an electrical board worker. The worker then manually verifies/checks and updates the status. If no theft is discovered, a bill is generated. Key Words: Electricity Theft, Machine Learning, Optical Character Recognition, SARIMAX

Modular Advanced Metering Infrastructure to Reduce Electricity Theft and a Cluster-Based Illegal Loads Detection

This article introduces the development of a modu- lar advanced metering infrastructure (AMI) for a single-phase power system, focusing on the reduction of non-technical losses (NTLs). In emerging economies, electromechanical meters are within the reach of consumers, and can be easily tampered with. With the proposed modular AMI, the physical security of smart meters is increased, thereby reducing this practice. The hardware design and the functionality of each main component of the modular AMI are explained. In addition, a cluster-based strategy to detect illegal electrical loads directly connected to power distribution lines is presented. The detection algorithm does not require extensive processing or complicated analysis of a large amount of data. The experimental and numerical results confirm the functionality of the developed AMI, an adequate detection of energy theft, and the feasibility to reduce the NTLs.

ELECTRICITY THEFT DETECTION USING MACHINE LEARNING

Non-technical losses (NTL) in electric power grids arise through electricity theft, broken electric meters or billing errors. They can harm the power supplier as well as the whole economy of a country through losses of up to 40% of the total power distribution. For NTL detection, researchers use artificial intelligence to analyse data. This work is about improving the extraction of more meaningful features from a data set. With these features, the prediction quality will increase.

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.

ELECTRICITY CONSUMERS AND ELECTRICITY THEFT IN EDO STATE, NIGERIA: IMPLICATION FOR EFFICIENT POWER DISTRIBUTION

Over the years, Nigeria has faced the challenges of inadequate generation, supply, and distribution of power. There is growing concern about the pervasiveness of electricity theft in the country, suggested to be among the chief factors inhibiting electricity sector efficiencies. The present study aimed to ascertain the trend of electricity theft among electricity consumers in Edo state, Nigeria. The study adopted a survey method to examine the prevalence of electricity theft among households in randomly selected towns in Edo state, Nigeria. The study population comprised electricity consumers under the residential category described by the BEDC. A convenient sample of 217 residential consumers and BEDC officials completed the research questionnaires. The finding shows that 32.7% of the electricity consumers indicated they never participated in energy theft, while 67.3% showed high involvement. A linear regression analysis was conducted to answer whether electricity theft affects efficient power distribution in Edo state. The result demonstrated a statistically significant effect of electricity theft on efficient power distribution in Edo state F (1, 12), 21.36 P< .05. Also, the R2 indicated that electricity theft contributed about 54.4% of the inefficiencies in power distribution in Edo state. Thus, it is concluded that the trend of electricity theft is high in the area of study and significantly affects the efficient distribution of electricity. The result provides valuable data to the government and electricity regulators.