Search

Non-Technical Losses (NTL) represent a serious concern for electric companies. These losses are responsible for revenue losses, as well as reduced system reliability. Part of the revenue loss is charged to legal consumers, thus, causing social imbalance. NTL methods have been developed in order to reduce the impact in physical distribution systems and legal consumers. These methods can be classified as hardware-based and non-hardware-based. Hardware-based methods need an entirely new system infrastructure to be implemented, resulting in high investment and increased cost for energy companies, thus hampering implementation in poorer nations. With this in mind, this paper performs a review of non-hardware-based NTL detection methods. These methods use distribution systems and consumers' data to detect abnormal energy consumption. They can be classified as network-based, which use network technical parameters to search for energy losses, data-based methods, which use data science and machine learning, and hybrid methods, which combine both. This paper focuses on reviewing non-hardware-based NTL detection methods, presenting a NTL detection methods overview and a literature search and analysis. [ABSTRACT FROM AUTHOR]

Non-technical losses (NTL) in power grid not only cause significant financial loss but also reduce the quality of supply for the distribution network operators (DNOs). In recent years, the frequency of power theft activities related to NTL have been increased rapidly. The data-based detection approach for NTLs has been studied by researchers from all over the world thoroughly, but the region domain shift issues and the lack of labelled data limited its applicability. In this paper, an adversarial learning-based approach is proposed to alleviate the domain shift problem, named Cross-Regional Adaptive Network (CAN). First of all, a multilayer Convolutional Neural Networks (CNN) feature extractor maps the training and application scenario data into the same feature space. After that the feature extractor and the domain classifier are trained against each other by inversing extractor's gradients. Finally, the training and application scenario data are indistinguishable for the domain classifier, and the feature extractor can extract domain invariant features. Experiments are conducted on two sets of electricity consumption data captured by utility company to verify the effectiveness of the proposed method. • It is found that the DA-EFD algorithm with domain adaptation improved 23.1%, 13.9% and 14.0% in accuracy, AUC and PR-AUC over EFD without domain adaptation. • DA-EFD algorithm can transfer model trained on labelled datasets to unlabelled data without any labelled information in the target domain, which may help to address the scarcity of labelled data. [ABSTRACT FROM AUTHOR]

Abstract: Finding an optimal subset of features that maximizes classification accuracy is still an open problem. In this paper, we exploit the speed of the Harmony Search algorithm and the Optimum-Path Forest classifier in order to propose a new fast and accurate approach for feature selection. Comparisons to some other pattern recognition and feature selection techniques showed that the proposed hybrid algorithm for feature selection outperformed them. The experiments were carried out in the context of identifying non-technical losses in power distribution systems. [Copyright &y& Elsevier]

• It is found that the proposed algorithm is 20% higher than existing methods. • The NAS algorithm achieved a higher accuracy with 1/10 of the calculating time. • No specific expertise required for NAS search method in building the model. • This allowed the DNO companies detect NTLs when their patterns evolve. The fast growth of non-technical loss (NTL) has gradually become one of the main concerns for distribution network operators (DNOs). Electricity theft which constitutes the main part of NTL not only brings losses to the DNOs, but also reduces the quality of the supply. A traditional detection method relies on utility workers' experience and consumes a large amount of manpower. Thanks to the emerging of advanced metering infrastructure (AMI), utility companies can now collect detailed data reflecting consumers' electricity usage, which enabled algorithms-based non-technical loss detection. The current data-based methods focus on the characteristics of electricity consumption thereby less efficient when dealing with rapidly changed electricity theft techniques. This article introduced a new data set, the location information of missing values, to improve the accuracy of non-technical loss detection. The relationship between missing values and electricity theft techniques is analyzed and a neural network model is built through neural architecture search (NAS). The improved model achieved an excellent Area Under Curve (AUC) value around 0.926 which verified the close link between missing values and electricity theft techniques. The nature of neural architecture search allows automatic model update which makes it a user-friendly tool even for engineers without any neural network expertise. A case study was carried out in which the missing value pattern was analyzed through Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering algorithm. [ABSTRACT FROM AUTHOR]

Utility companies have a great interest in identifying energy losses. Here, we focus on Non-Technical Losses (NTL), which refer to losses caused by utility theft or meter errors. Typically, utility companies resort to machine learning solutions to automate and optimise the identification of such losses. This paper extends an existing NTL-detection framework: by including knowledge-based NTL segmentation, we have detected some opportunities for improving the accuracy and the explanations provided to the utility company. Our improved models focus on specific types of NTL and therefore, the explanations provided are easier to interpret, allowing stakeholders to make more informed decisions. The improvements and results presented in the article may benefit other industrial frameworks. [ABSTRACT FROM AUTHOR]

• A practical solution to curb the electricity theft occurring due to hook-line activity. • The effectiveness, ease of application and flexibility of proposed electricity theft prevention technique is tested for various legal and illegal loads. • This technique curbs the non-technical losses, balances the supply demand response and reduces additional tariff posed on the genuine consumers. • It restricts the overloading of DT by preventing hook-line activity. Thus, life expectancy of DT will increase. The pilferage of electricity is a major reason for high aggregate technical and commercial losses. New technologies and methodologies incorporating anti-tampering features have been developed in metering system to arrest pilferage, but innovative means have been devised by the unscrupulous elements of society to counter the same, sometimes in connivance with those who are expected to check such activities. The proposed energy theft prevention scheme is an online approach which performs theft detection and prevention in a real-time scenario without any supply interruption. The aim is accomplished by design and implementation of electricity pilferage prevention system (EPPS) developed to provide a different amplitude of unsuitable supply voltage for random time intervals to illegally connected loads depending on the voltage regulation of power distribution network. In order to maintain uninterrupted supply to genuine consumers, consumer supervision unit (CSU) is provided by the utility. Thus, consumer appliances will operate efficiently as the voltage regulation is improved by preventing hook-line activity. The proposed theft prevention scheme benefits both utility and consumer in terms of increase in revenue and reliability of power supply respectively. The suitability, flexibility in operation and effectiveness of proposed EPPS and CSU based theft prevention scheme is experimentally demonstrated for different types of legal and illegal consumer loads under various voltage regulation scenario by using a microcontroller model STM32L151x6/8/B. In our experiments the electricity theft detects on the basis of users' data with 99% accuracy. Also the theft prevention in a real-time scenario without any supply interruption and without knowing the exact topology of the power distribution network is operated with 100% accuracy. [ABSTRACT FROM AUTHOR]

• Ranking of customers with the probability of having non-technical losses. • Detection and location of energy thefts by hypothesis tests of error estimates. • Detection of non-technical losses caused by energy thefts. • State estimation application for electrical distribution systems. • Use of AMI for state estimation and energy theft detection. Non-technical losses are an important issue for distribution utilities worldwide, and energy thefts are among their main causes. To mitigate this problem, utilities must detect and remove irregular connections from their networks. A traditional practice is to make local inspections guided by the results of artificial intelligence techniques relying on data from previous inspections and socio-economic features. However, the success rate still tends to be low and massive inspections of customers are expensive. Therefore, utilities require methods that improve their inspection success rate. In this context, this paper proposes a method to identify and rank non-technical losses in distribution systems. It uses a hypothesis test based on state estimation error distributions. By using the method, the probability of having a non-technical loss can be computed for every customer, which is improved by the method's repetition over time. By doing this, customers suspected of non-technical loss are ranked. Tests carried out on a real feeder have shown that the method is promising for locating multiple non-technical losses in low-voltage networks. In the tests, the minimum theft power to be identified is around 400 W. However, the location has good performance for power of 1 kW and above. Besides guiding utilities' inspection teams on non-technical loss detection, the proposal represents a way to benefit from advanced measurement infrastructure that has been installed in distribution networks worldwide. [ABSTRACT FROM AUTHOR]

Artificial intelligence techniques lead to data-driven energy services in distribution power systems by extracting value from the data generated by the deployed metering and sensing devices. This paper performs a holistic analysis of artificial intelligence applications to distribution networks, ranging from operation, monitoring and maintenance to planning. The potential artificial intelligence techniques for power system applications and needed data sources are identified and classified. The following data-driven services for distribution networks are analyzed: topology estimation, observability, fraud detection, predictive maintenance, non-technical losses detection, forecasting, energy management systems, aggregated flexibility services and trading. A review of the artificial intelligence methods implemented in each of these services is conducted. Their interdependencies are mapped, proving that multiple services can be offered as a single clustered service to different stakeholders. Furthermore, the dependencies between the AI techniques with each energy service are identified. In recent years there has been a significant rise of deep learning applications for time series prediction tasks. Another finding is that unsupervised learning methods are mainly being applied to customer segmentation, buildings efficiency clustering and consumption profile grouping for non-technical losses detection. Reinforcement learning is being widely applied to energy management systems design, although more testing in real environments is needed. Distribution network sensorization should be enhanced and increased in order to obtain larger amounts of valuable data, enabling better service outcomes. Finally, the future opportunities and challenges for applying artificial intelligence in distribution grids are discussed. • Identification of main data-driven techniques and data sources in power systems. • Proposal and review of data-driven energy services and how they interrelate. • The dependencies among AI techniques and data-driven energy services are mapped. • Opportunities and challenges regarding the distribution grid digitalization. [ABSTRACT FROM AUTHOR]

Non-technical losses (NTL) identification has been paramount in the last years. However, it is not straightforward to obtain labelled datasets to perform a supervised NTL recognition task. In this paper, the optimum-path forest (OPF) clustering algorithm has been employed to identify irregular and regular profiles of commercial and industrial consumers obtained from a Brazilian electrical power company. Additionally, a model for the problem of NTL recognition as an anomaly detection task has been proposed when there are little or no information about irregular consumers. For such purpose, two new approaches based on the OPF framework have been introduced and compared against the well-known k -means, Gaussian mixture model, Birch, affinity propagation and one-class support vector machines. The experimental results have shown the robustness of OPF for both unsupervised NTL recognition and anomaly detection problems. In short, the main contributions of this paper are fourfold: (i) to employ unsupervised OPF for non-technical losses detection, (ii) to model the problem of NTL as being an anomaly detection task, (iii) to employ unsupervised OPF to estimate the parameters of the Gaussian distributions, and (iv) to present an anomaly detection approach based on unsupervised optimum-path forest. [ABSTRACT FROM AUTHOR]