Comparativo de técnicas de aprendizado de máquinas para a previsão de faltas elétricas
DOI:
https://doi.org/10.53660/362.prw907dPalavras-chave:
Redes de Distribuição, Aprendizado de Maquinas, Predição Numérica de Tempo, Faltas ElétricasResumo
Neste estudo é elaborado um modelo de previsão de faltas na rede de distribuição de energia no estado do Espírito Santo, Brasil, demonstrando quais regiões têm mais possibilidades de serem afetadas. Para isso, serão utilizados dados históricos de falhas, previsão numérica do tempo, cobertura terrestre e infraestrutura da rede. Essas informações são combinadas em regiões com o sistema de geocódigo Geohash e agrupadas por dia, servido assim de entrada para o modelo de aprendizado de máquina como Decision Tree (DT), Randon Florest(RF), K-Nearest Neighbors (KNN) e Support Vector Machine (SVM). Os resultados indicaram que a abordagem proposta apresentou uma precisão significativa na identificação da localização de falta, apresentando uma acurácia de 0.707 e um erro quadrático médio de 0.293.
Downloads
Referências
MEHL, Ewaldo LM. Qualidade da energia elétrica. UNIVERSIDADE FEDE-RAL DO PARANÁ–UFPR, p. 21, 2012.
DE DISTRIBUIÇÃO, Prodist—Procedimentos. Módulo 8–qualidade da energia elétrica. Agência Nacional de Energia Elétrica–ANEEL, 2010.
WARD, David M. The effect of weather on grid systems and the reliability of electricity supply. Climatic Change, v. 121, n. 1, p. 103-113, 2013.
MCGUFFIE, K.; HENDERSON‐SELLERS, A. Forty years of numerical climate modelling. International Journal of Climatology: A Journal of the Royal Meteorological Society, v. 21, n. 9, p. 1067-1109, 2001.
BENDRE, M. R.; THOOL, R. C.; THOOL, V. R. Big data in precision agriculture: Weather forecasting for future farming. In: 2015 1st International Conference on Next Generation Computing Technologies (NGCT). IEEE, 2015. p. 744-750.
BAKLANOV, Alexander et al. Potential and shortcomings of numerical weather prediction models in providing meteorological data for urban air pollution forecasting. Water, Air and Soil Pollution: Focus, v. 2, p. 43-60, 2002.
BALKIĆ, Zoran; ŠOŠTARIĆ, Damir; HORVAT, Goran. GeoHash and UUID identifier for multi-agent systems. In: Agent and Multi-Agent Systems. Technologies and Applications: 6th KES International Conference, KES-AMSTA 2012, Dubrovnik, Croatia, June 25-27, 2012. Proceedings 6. Springer Berlin Heidelberg, 2012. p. 290-298.
SOUSA, Roniel S. De; BOUKERCHE, Azzedine; LOUREIRO, Antonio AF. Vehicle trajectory similarity: models, methods, and applications. ACM Computing Surveys (CSUR), v. 53, n. 5, p. 1-32, 2020.
ALTOWAYAN, A. Aziz; TAO, Lixin. Word embeddings for Arabic sentiment analysis. In: 2016 IEEE International Conference on Big Data (Big Data). IEEE, 2016. p. 3820-3825.
LORENZ, Elke et al. Irradiance forecasting for the power prediction of grid-connected photovoltaic systems. IEEE Journal of selected topics in applied earth observations and remote sensing, v. 2, n. 1, p. 2-10, 2009.
ODOM, Marcus D.; SHARDA, Ramesh. A neural network model for bankruptcy prediction. In: 1990 IJCNN International Joint Conference on neural networks. IEEE, 1990. p. 163-168.
TASKINEN, Jyrki; YLIRUUSI, Jouko. Prediction of physicochemical properties based on neural network modelling. Advanced drug delivery reviews, v. 55, n. 9, p. 1163-1183, 2003.
LOPES, Ludmila Junca. Sistema inteligente de apoio a decisão para a fase rampa de aceleração do processo de partida a quente de poços de elevação BCSS. 2017.
KONDO, Daniela Vinci. Alocação de religadores automatizados em sistemas de distribuição. 2015. Tese de Doutorado. Universidade de São Paulo.
CERRAI, Diego et al. Predicting storm outages through new representations of weather and vegetation. IEEE Access, v. 7, p. 29639-29654, 2019.
HERSBACH, Hans et al. ERA5 hourly data on single levels from 1979 to present. Copernicus climate change service (c3s) climate data store (cds), v. 10, n. 10.24381, 2018.
LI, Jin; HEAP, Andrew D. A review of spatial interpolation methods for environmental scientists. 2008.
RAHMAN, KM Jawadur; MUNNEE, Maria Moosa; KHAN, Shahriar. Largest blackouts around the world: Trends and data analyses. In: 2016 IEEE International WIE Conference on Electrical and Computer Engineering (WIECON-ECE). IEEE, 2016. p. 155-159.
FERNANDO, Malith; MILLANGODA, Malinda; PREMALAL, Sarath. Analyze and Comparison of the Atmospheric Instability Using K-Index, Lifted Index Total Totals Index Convective Availability Potential Energy (CAPE) and Convective Inhibition (CIN) in Development of Thunderstorms in Sri Lanka During Second Inter-Monsoon. In: Multi-Hazard Early Warning and Disaster Risks. Springer International Publishing, 2021. p. 603-614.
HOU, Hui et al. Review of failure risk and outage prediction in power system under wind hazards✰. Electric Power Systems Research, v. 210, p. 108098, 2022.
ZANAGA, Daniele et al. ESA WorldCover 10 m 2021 v200. 2022.
LAROSE, Daniel T. Data mining and predictive analytics. John Wiley & Sons, 2015.
HASTIE, Trevor et al. The elements of statistical learning: data mining, inference, and prediction. New York: springer, 2009.
DOMINGOS, Pedro. A few useful things to know about machine learning. Communications of the ACM, v. 55, n. 10, p. 78-87, 2012.
CHOI, Hosik; PARK, Hye Won; PARK, Changyi. Support vector machines for big data analysis. Journal of the Korean Data and Information Science Society, v. 24, n. 5, p. 989-998, 2013.
GOODFELLOW, Ian; BENGIO, Yoshua; COURVILLE, Aaron. Deep learning. MIT press, 2016.
