Probabilistic models for commercial losses detection

• Grant number: 17/02286-0
• Support Opportunities: Regular Research Grants
• Start date: July 01, 2017
• End date: June 30, 2019
• Field of knowledge: Engineering - Electrical Engineering - Power Systems
• Principal Investigator: André Nunes de Souza
• Grantee: André Nunes de Souza
• Host Institution: Faculdade de Engenharia (FE). Universidade Estadual Paulista (UNESP). Campus de Bauru. Bauru , SP, Brazil
• Associated researchers: João Paulo Papa ; Pedro da Costa Junior

Abstract

Currently, non-technical losses detection is an active and a very concerning problem, mainly in development countries such as Brazil, where the amount of illegal connections may reach worrying levels. Such connections can somehow compromise the lifespan of devices usually employed in the energy transmission and transformation, since they were firstly designed to work under a certain amount of known users. This proposal aims at employing probabilistic models in the context of fraud detection in domestic and industrial energy systems. Such techniques can model the dynamics of data patterns, thus being more efficient in datasets that can change overtime, such as energy users. Therefore, the main idea of this project would be to predict when a certain customer could become illegal, and not only to just identify him. This proposal also comprises researchers from different fields, as well as graduate students. (AU)

Scientific publications

(10)

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

PASSOS, LEANDRO A.; PAPA, JOAO PAULO; HUSSAIN, AMIR; ADEEL, AHSAN. Canonical cortical graph neural networks and its application for speech enhancement in audio-visual hearing aids. Neurocomputing, v. 527, p. 8-pg., 2023-01-20. (14/12236-1, 19/07665-4, 13/07375-0, 17/02286-0, 19/18287-0, 18/21934-5)

DE ROSA, GUSTAVO H.; PAPA, JOAO P.; YANG, XIN-SHE. A nature-inspired feature selection approach based on hypercomplex information. APPLIED SOFT COMPUTING, v. 94, SEP 2020. (13/07375-0, 16/19403-6, 14/12236-1, 17/25908-6, 17/02286-0, 19/02205-5)

JODAS, DANILO SAMUEL; PASSOS, LEANDRO APARECIDO; ADEEL, AHSAN; PAPA, JOAO PAULO. PL-kNN: A Python-based implementation of a parameterless k-Nearest Neighbors classifier. SOFTWARE IMPACTS, v. 15, p. 3-pg., 2022-12-21. (14/12236-1, 19/07665-4, 13/07375-0, 17/02286-0, 19/18287-0, 18/21934-5)

AFONSO, LUIS C. S.; PEDRONETTE, DANIEL C. G.; DE SOUZA, ANDRE N.; PAPA, JOAO P.; IEEE. Improving Optimum-Path Forest Classification Using Unsupervised Manifold Learning. 2018 24TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), v. N/A, p. 6-pg., 2018-01-01. (14/12236-1, 13/07375-0, 17/22905-6, 13/08645-0, 17/02286-0, 16/19403-6)

RODRIGUES, DOUGLAS; DE ALBUQUERQUE, VICTOR HUGO C.; PAPA, JOAO PAULO. A multi-objective artificial butterfly optimization approach for feature selection. APPLIED SOFT COMPUTING, v. 94, SEP 2020. (14/12236-1, 14/16250-9, 17/02286-0, 16/19403-6)

PASSOS, LEANDRO APARECIDO S.; JODAS, DANILO S.; RIBEIRO, LUIZ C. F.; AKIO, MARCO; DE SOUZA, ANDRE NUNES; PAPA, JOAO PAULO. Handling imbalanced datasets through Optimum-Path Forest. KNOWLEDGE-BASED SYSTEMS, v. 242, p. 13-pg., 2022-04-22. (18/21934-5, 14/12236-1, 20/12101-0, 19/18287-0, 19/07665-4, 17/02286-0, 13/07375-0)

IWASHITA, ADRIANA SAYURI; RODRIGUES, DOUGLAS; GASTALDELLO, DANILO SINKITI; DE SOUZA, ANDRE NUNES; PAPA, JOAO PAULO. An incremental Optimum-Path Forest classifier and its application to non-technical losses identification. COMPUTERS & ELECTRICAL ENGINEERING, v. 95, OCT 2021. (14/12236-1, 19/07665-4, 18/21934-5, 17/02286-0, 13/07375-0)

PAPA, JOAO P.; ROSA, GUSTAVO H.; DE SOUZA, ANDRE N.; AFONSO, LUIS C. S.. Feature selection through binary brain storm optimization. COMPUTERS & ELECTRICAL ENGINEERING, v. 72, p. 468-481, NOV 2018. (13/07375-0, 17/02286-0, 16/19403-6, 14/12236-1, 17/22905-6, 13/08645-0)

FERNANDES, SILAS E. N.; PEREIRA, DANILLO R.; RAMOS, CAIO C. O.; SOUZA, ANDRE N.; GASTALDELLO, DANILO S.; PAPA, JOAO P.. A Probabilistic Optimum-Path Forest Classifier for Non-Technical Losses Detection. IEEE TRANSACTIONS ON SMART GRID, v. 10, n. 3, p. 3226-3235, MAY 2019. (13/07375-0, 14/16250-9, 17/02286-0, 16/19403-6, 14/12236-1)

JODAS, DANILO SAMUEL; PASSOS, LEANDRO APARECIDO; ADEEL, AHSAN; PAPA, JOAO PAULO; IEEE. PL-kNN: A Parameterless Nearest Neighbors Classifier. 2022 29TH INTERNATIONAL CONFERENCE ON SYSTEMS, SIGNALS AND IMAGE PROCESSING (IWSSIP), v. N/A, p. 4-pg., 2022-01-01. (14/12236-1, 19/07665-4, 17/02286-0, 19/18287-0, 18/21934-5)