Non-technical Losses Identification Research Articles

Operational Identification of Resistances of Wires of 380 V Distribution Networks by Automated Accounting Systems

Four-wire distribution electrical networks of 0.4 kV equipped with automated systems of electrical energy accounting are considered. The problem of identifying the resistances of the wires of the distribution network is solved on the basis of the effective values of voltages and currents, as well as of their phase shift angles obtained by the accounting system in the power supply node of the network and from its subscribers for the selected observation intervals. A brief analysis of the known methods and technologies used in this area of research is carried out. The importance of the formulated problem for applied problems, such as control and diagnostics of electrical energy losses, as well as the technical condition of the network, is noted. A method (algorithm) is proposed that allows determining unknown, unequal complex resistances of inter-subscriber sections of the distribution network. In these resistances, the reactive components are considered equal within the inter-subscriber section; the active components differ due to the influence of unequal flowing currents and/or weather factors. At the same time, data from two different network operating modes are required, which are selected based on the analysis of the dynamics of changes of supply currents and/or voltages by the accounting system device connected to the power supply node. Considering that the active resistances of the wires must remain unchanged, the mode that is used for calculations is the one that is before the change in power consumption in the network and the next one immediately (about 0.1 s) after it. An example of a calculation that demonstrates the reliability of the proposed equations of the method that has been developed on a simulated distribution network is given. The research results are focused on the improvement of automated accounting systems and the implementation of their new functions that elevate the reliability of distribution networks, as well as allowing for the rapid identification of non-technical losses of electrical energy.

Identification of Nontechnical Losses in Distribution Systems Adding Exogenous Data and Artificial Intelligence

Nontechnical losses (NTL) are irregularities in the consumption of electricity and mainly caused by theft and fraud. NTLs can be characterized as outliers in historical data series. The use of computational tools to identify outliers is the subject of research around the world, and in this context, artificial neural networks (ANN) are applicable. ANNs are machine learning models that learn through experience, and their performance is associated with the quality of the training data together with the optimization of the model’s architecture and hyperparameters. This article proposes a complete solution (end-to-end) using the ANN multilayer perceptron (MLP) model with supervised classification learning. For this, data mining concepts are applied to exogenous data, specifically the ambient temperature, and endogenous data from energy companies. The association of these data results in the improvement of the model’s input data that impact the identification of consumer units with NTLs. The test results show the importance of combining exogenous and endogenous data, which obtained a 0.0213 improvement in ROC-AUC and a 6.26% recall (1).

Identification of Non-technical Losses in Smart Metering Systems via a Dispersive Flies Algorithm

Identification of Non-technical Losses in Smart Metering Systems via a Dispersive Flies Algorithm

An incremental Optimum-Path Forest classifier and its application to non-technical losses identification

Non-technical losses stand for the energy consumed but not billed, affecting the energy grid as a whole. Such an issue somehow prevails in developing countries, harming the quality of energy and preventing social programs benefit from tax revenues. Machine learning techniques can help mitigate it by mining information from fraudsters and legal users for further decision-making. In this paper, we deal with a steady increase of dataset size, i.e., the incremental learning problem, which can cope with datasets regularly provided by energy companies, requiring the learner to be updated constantly. Since repeating the entire learning process might be prohibitive, adjusting the model to the new data shows to be a better choice. We propose an incremental Optimum-Path Forest approach with k-nn neighborhood that is considerably more efficient for training than its counterpart version, with experiments validated in general-purpose datasets and also in the context of non-technical losses identification.

Review of Non-Technical Losses Identification Techniques

Illegally consumption of electric power, termed as non-technical losses for the distribution companies is one of the dominant factors all over the world for many years. Although there are some conventional methods to identify these irregularities, such as physical inspection of meters at the consumer premises etc, but it requires large number of manpower and time; then also it does not seem to be adequate. Now a days there are various methods and algorithms have been developed that are proposed in different research papers, to detect non-technical losses. In this paper these methods are reviewed, their important features are highlighted and also the limitations are identified. Finally, the qualitative comparison of various non-technical losses identification algorithms is presented based on their performance, costs, data handling, quality control and execution times. It can be concluded that the graph-based classifier, Optimum-Path Forest algorithm that have both supervised and unsupervised variants, yields the most accurate result to detect non-technical losses.

A Probabilistic Optimum-Path Forest Classifier for Non-Technical Losses Detection

Probabilistic-driven classification techniques extend the role of traditional approaches that output labels (usually integer numbers) only. Such techniques are more fruitful when dealing with problems where one is not interested in recognition/identification only, but also into monitoring the behavior of consumers and/or machines, for instance. Therefore, by means of probability estimates, one can take decisions to work better in a number of scenarios. In this paper, we propose a probabilistic-based optimum-path forest (OPF) classifier to handle the problem of non-technical losses (NTL) detection in power distribution systems. The proposed approach is compared against naive OPF, probabilistic support vector machines, and logistic regression, showing promising results for both NTL identification and in the context of general-purpose applications.

Nonintrusive Energy Meter for Nontechnical Losses Identification

We present in this paper a method and apparatus for nonintrusive measurement of active energy in low-voltage ac installations. In the proposed method, the active power is calculated from the voltage and current waveforms, and the phase shift between them. No voltage amplitude measurement is required since the nominal voltage of the ac installation is considered as the actual voltage. This approach bypasses the main disadvantage of capacitive ac voltage probing: low accuracy in amplitude measurements due to variations on the sensor coupling capacitance. The energy meter that implements the technique is composed of a commercial nonintrusive current sensor (clamp-on current transformer) and a contactless capacitive voltage sensor—designed to measure the voltage phase and waveform. The voltage sensor includes a shield to isolate it from external electric fields, making the energy meter suitable for applications in polyphase systems. The developed energy meter does not require on-site calibration, galvanic contact to the phase conductors, neither electrical circuit opening, allowing for quick, safe, and easy installation on overhead service drop line. Due to these characteristics, it can be used by electricity distribution companies in the preinspection of consumer units suspected of fraud. Experimental results proved that the developed energy meter is insensitive to the characteristics of cables (width and insulation), external electrical fields, as well as to the voltage sensor capacitance variations. The error of the active energy measurement under real condition in a two-phase installation was 1%.

Clustering-based novelty detection for identification of non-technical losses

The reduction of non-technical losses is a significant part of the total potential benefits resulting from implementations of the smart grid concept. This paper proposes a data-based method to detect sources of theft and other commercial losses. Prototypes of typical consumption behavior are extracted through clustering of data collected from smart meters. A distance-based novelty detection framework classifies new data samples as malign if their distance to the typical consumption prototypes is significant. The proposed method works on the space of four different indicators of irregular consumption, enabling the easy interpretation of results. A use case based on real data is presented to evaluate the method. The threat model considers sixteen different possible types of changes in consumption pattern that result from non-technical losses, including attacks and defects present since the first day of metering. The proposed clustering-based novelty detection method for identification of non-technical losses, using the Gustafson-Kessel fuzzy clustering algorithm, achieves a true positive rate of 63.6% and false positive rate of 24.3%, outperforming other state-of-the-art unsupervised learning methods.

Unsupervised non-technical losses identification through optimum-path forest

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.