Fast Non-Technical Losses Identification Through Optimum-Path Forest (original) (raw)
Related papers
Artificial intelligence for energy fraud detection: a review
International Journal of Applied Power Engineering (IJAPE)
Energy fraud in the distribution sector of electric utility includes electricity theft, meter tampering, or billing error. This fraud causing non-technical loss has led to an economic loss of the company. In order to detect and minimize fraud, different technologies have been used. From conventional methods to development in the field of artificial intelligence (AI), effective and reliable fraud detection methods have been proposed. This paper first provides an overview of different proposed methods for non-technical loss detection and evaluate the advantage and limitation of using those methods. Furthermore, several supervised and unsupervised machine learning methods for detecting electricity theft are discussed in summary along with their metrics and attributes used. Finally, these methods are classified based on the overall operation and the parameters used. This paper provides comparisons of several fraud detection methods using AI along with their weak and strong points and th...
A framework to select a classification algorithm in electricity fraud detection
South African Journal of Science, 2020
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.
International Journal of Electrical Power & Energy Systems, 2012
For the electrical sector, minimizing non-technical losses is a very important task because it has a high impact in the company profits. Thus, this paper describes some new advances for the detection of non-technical losses in the customers of one of the most important power utilities of Spain and Latin America: Endesa Company. The study is within the framework of the MIDAS project that is being developed at the Electronic Technology Department of the University of Seville with the funding of this company. The advances presented in this article have an objective of detecting customers with anomalous drops in their consumed energy (the most-frequent symptom of a non-technical loss in a customer) by means of a windowed analysis with the use of the Pearson coefficient. On the other hand, besides Bayesian networks, decision trees have been used for detecting other types of patterns of non-technical loss. The algorithms have been tested with real customers of the database of Endesa Company. Currently, the system is in operation.
Electricity consumer fraud is a problem faced by all power utilities. Finding efficient measurements for detecting fraudulent electricity consumption has been an active research area in recent years. In this paper,the approach towards nontechnical loss (NTL) detection in power utilities using an artificial intelligence based technique, Support Vector Machine (SVM), are presented. This approach provides a method of data mining, which involves feature extraction from past consumption data. This SVM based approach uses customer load profile information and additional attributes to expose abnormal behavior that is known to be highly correlated with NTL activities. Some key advantages of SVM in data clustering, among which is the easy way of using them to fit the data of a wide range of features are discussed here. Finally, some major weakness of using SVM in clustering for NTL identification are identified, which leads to motivate for the scope of Optimum-Path Forest, a new model of NTL identification.
Non-technical losses: detection methods and regulatory aspects overview
CIRED - Open Access Proceedings Journal
Non technical losses have been reported as one of the most serious problems faced by Electric Utilities. This paper provides an overview of most recent methods for electricity fraud detection, based on concepts from data mining, state estimation, game theory etc. 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.
Increasing the efficiency in Non-Technical Losses detection in utility companies
Melecon 2010 - 2010 15th IEEE Mediterranean Electrotechnical Conference, 2010
Usually, the fraud detection method in utility companies uses the consumption information, the economic activity, the geographic location, the active/reactive ration and the contracted power. This paper proposes a combined text mining and neural networks to increase the efficiency in Non-Technical Losses (NTLs) detection methods which was previously applied. This proposed framework proposes to collect all the information that normally cannot be treated with traditional methods. This framework is part of a research project. This project is done in collaboration with Endesa, one of the most important power distribution companies of Europe. Currently, the proposed framework is in the test stage and it uses real cases.
Mathematics
Several approaches have been proposed to detect any malicious manipulation caused by electricity fraudsters. Some of the significant approaches are Machine Learning algorithms and data-based methods that have shown advantages compared to the traditional methods, and they are becoming predominant in recent years. In this study, a novel method is introduced to detect the fraudulent NTL loss in the smart grids in a two-stage detection process. In the first stage, the time-series readings are enriched by adding a new set of extracted features from the detection of sudden Jump patterns in the electricity consumption and the Autoregressive Integrated moving average (ARIMA). In the second stage, the distributed random forest (DRF) generates the learned model. The proposed model is applied to the public SGCC dataset, and the approach results have reported 98% accuracy and F1-score. Such results outperform the other recently reported state-of-the-art methods for NTL detection that are applie...
Advances in Multidisciplinary and scientific Research Journal Publication, 2021
Over the years, Electricity theft has been estimated to cost billions of Naira per year in Nigeria. To reduce electricity theft, electric utilities are leveraging data collected by using data analytics to identify abnormal consumption trends and possible fraud. In this study, use of data analytics in detecting electricity theft, and a metric that leverages this threat model in order to evaluate and compare anomaly detectors. Data mining technology have helped several industries and sectors in improving their various forms of technology, this study therefore employ machine learning algorithms for the classification of fraud detection in the electricity consumption of costumers. In this project, Support Vector Machine (SVM) and C4.5 Decision tree classification algorithms were employed for fraud detection using customer electricity consumption data. SVM and C4.5 achieved 63.4% and 65.9% accuracy respectively. Thus, C4.5 outperformed SVM based on the dataset experimented in this projec...