Social-Spider Optimization-based Support Vector Machines applied for energy theft detection

Pereira, Danillo Roberto; Pazoti, Mário Augusto; Pereira, Luiz Haroldo; Rodrigues, Douglas; Ramos, Caio O.; Souza, Ana Cristina de; Papa, João Paulo

doi:10.1016/j.compeleceng.2015.11.001

Cited by 55 publications

(26 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In order to study the behavior of P-OPF under different scenarios, we used three synthetic datasets (synhetic0, synhetic2, synhetic3), two datasets concerning energy theft detection (comercial and industrial) [17], as well as nine public benchmarking datasets 2 . These datasets have been frequently used in the evaluation of different classification methods.…”

Section: Methodsmentioning

confidence: 99%

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

Fernandes

Pereira

Ramos

et al. 2019

IEEE Trans. Smart Grid

Self Cite

View full text Add to dashboard Cite

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 with binary classification problems, and we show it can be more accurate than naïve OPF in a number of datasets. In addition to being just more accurate or not, probabilistic OPF turns to be another useful tool to the scientific community.

show abstract

Section: Methodsmentioning

confidence: 99%

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

Fernandes

Pereira

Ramos

et al. 2019

IEEE Trans. Smart Grid

Self Cite

View full text Add to dashboard Cite

show abstract

“…The detection of irregular users through computational intelligence has been treated by many authors with several approaches. The development of intelligent systems has been an alternative and such systems have included techniques as artificial neural networks (ANNs) (Markoč, Hlupić, & Basch, 2011;Zheng, Yang, Niu, Dai, & Zhou, 2018), principle component analysis (PCA) (Singh, Bose, & Joshi, 2017), fuzzy models (Viegas, & Viera, 2017;Nagi et al, 2011), data mining (Chen et al, 2014) and support vector machines (SVMs) (Nagi J. et al, 2010;Pereira et al, 2016).…”

Section: Uparela Gonzalez Jimenez and Quinteromentioning

confidence: 99%

Intelligent system for non-technical losses management in residential users of the electricity sector

Cantillo¹,

González²,

Mares³

et al. 2018

Ing. Inv.

View full text Add to dashboard Cite

The identification of irregular users is an important assignment in the recovery of energy in the distribution sector. This analysis requires low error levels to minimize non-technical electrical losses in power grid. However, the detection of fraudulent users who have billing does not present a generalized methodology. This issue is complex and varies according to the case study. This paper presents a novel methodology to identify residential fraudulent users by using intelligent systems. The proposed intelligent system consists of three fundamental modules. The first module performs the classification of users with similar power consumption curves using self-organizing maps and genetic algorithms. The second module allows carrying out the monthly electricity demand forecasting through of recursive adjustment of ARIMA models. The third module performs the detection of fraudulent users through an artificial neural network for pattern recognition. For the design and validation of the proposed intelligent system, several tests were performed in each developed module. The database used for the design and evaluation of the modules was constructed with data supplied by the energy distribution company of the Colombian Caribbean Region. The results obtained by the proposed intelligent system show a better performance versus the detection rates obtained by the company. RESUMENLa identificación de usuarios con consumo fraudulento es una actividad importante en la recuperación de energía en el sector de la distribución. Este análisis requiere bajos niveles de error para minimizar las pérdidas eléctricas no técnicas en la red de distribución. Sin embargo, la detección de usuarios fraudulentos con facturación no tiene una metodología generalizada. Este es un problema complejo y varía de acuerdo con cada caso de estudio. Este artículo presenta una nueva metodología para la identificación inteligente de usuarios fraudulentos residenciales basada en sistemas inteligentes. El sistema inteligente propuesto consiste en tres módulos fundamentales. El primer módulo clasifica a los usuarios con curvas de consumo similares a través de mapas auto-organizativos y algoritmo genéticos. El segundo módulo realiza la predicción de consumos mensuales mediante ajustes recursivos de modelos ARIMA. El tercer módulo es el responsable de llevar a cabo la detección de usuarios irregulares por medio de una red neuronal para reconocimiento de patrones. Para el diseño y validación del sistema inteligente propuesto se realizaron pruebas en cada módulo que lo integra para diferentes tipos de clientes del mercado. La base de datos utilizada para el diseño y evaluación de los módulos fue construida a partir de los datos suministrados por la empresa de distribución de energía de la Costa Caribe Colombiana. Los resultados obtenidos por el sistema inteligente propuesto muestran un mejor desempeño frente a los índices de detección obtenidos por la empresa.Palabras clave: pérdidas no técnicas, consumo irregular de electricidad, detección de fraudes, sistemas i...

show abstract

“…Unlike other popular algorithms as Particle Swarm Optimization (PSO) [4], Genetic Algorithm (GA) [5], Cuckoo Search (CS) [6], Artificial Bee Colony (ABC) [7], Harmony Search (HS) [8], and Social Network Optimization (SNO) [9], it evades the concentration of particles in the best positions, preventing critical faults as premature convergence to suboptimal solutions or a limited balance between exploration and exploitation. These characteristics have motivated the use of the SSO algorithm to solve a wide variety of engineering applications in diverse areas including machine learning: Artificial Neural Networks Training [10,11] and Support Vector Machine Parameter Tuning [12,13]; control: Fractional Controller Design [14][15][16] and frequency controllers [17]; image processing: Image Multilevel Thresholding [18], Image Contrast Enhancement [19], and Image Template Matching [20], and energy: Distribution of Renewable Energy [21], Congestion Management [22], and Anti-Islanding Protection [23].…”

Section: Introductionmentioning

confidence: 99%

Social Spider Optimization Algorithm: Modifications, Applications, and Perspectives

Luque−Chang

Cuevas

Fausto

et al. 2018

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Swarm intelligence (SI) is a research field which has recently attracted the attention of several scientific communities. An SI approach tries to characterize the collective behavior of animal or insect groups to build a search strategy. These methods consider biological systems, which can be modeled as optimization processes to a certain extent. The Social Spider Optimization (SSO) is a novel swarm algorithm that is based on the cooperative characteristics of the social spider. In SSO, search agents represent a set of spiders which collectively move according to the biological behavior of the colony. In most of SI algorithms, all individuals are modeled considering the same properties and behavior. In contrast, SSO defines two different search agents: male and female. Therefore, according to the gender, each individual is conducted by using a different evolutionary operation which emulates its biological role in the colony. This individual categorization allows reducing critical flaws present in several SI approaches such as incorrect exploration-exploitation balance and premature convergence. After its introduction, SSO has been modified and applied in several engineering domains. In this paper, the state of the art, improvements, and applications of the SSO are reviewed.

show abstract

Social-Spider Optimization-based Support Vector Machines applied for energy theft detection

Cited by 55 publications

References 13 publications

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

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

Intelligent system for non-technical losses management in residential users of the electricity sector

Social Spider Optimization Algorithm: Modifications, Applications, and Perspectives

Contact Info

Product

Resources

About