Banca de DEFESA: MATEUS GUILHERME MELO DE SOUZA

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
STUDENT : MATEUS GUILHERME MELO DE SOUZA
DATE: 17/12/2020
TIME: 16:00
LOCAL: Remoto
TITLE:

DEVELOPMENT OF APPLICATION FOR OPERATION AND MAINTENANCE OF WIND TURBINES USING ARTIFICIAL INTELLIGENCE.

KEY WORDS:

Artificial Intelligence, Wind Turbine, Operation and Maintenance, Systematic Literature Review, ANN, LSTM.

PAGES: 97
BIG AREA: Engenharias
AREA: Engenharia de Produção
SUBÁREA: Engenharia do Produto
SPECIALTY: Desenvolvimento de Produto
SUMMARY:

Global warming has alerted the international community to the need to of renewable and clean energy sources in the generation of electricity. In this scenario, wind energy is expanding. As this source depends on a central equipment, the wind turbine, its operation and maintenance are fundamental for the viability of the business and, therefore, the application of technologies, such as Artificial Intelligence, can improve the competitiveness of the sector. The objective of this study is to present a framework for the application of Artificial Intelligence in the operation and maintenance of wind farms. For this, theoretical and field research were carried out. The theoretical research complemented a traditional literature review and a systematic literature review. State-of-the-art was identified through the analysis of 51 articles obtained from the Periodicos Capes Platform. The research identified the equipment studied, data, methods and metrics adopted in the application of AI. The field research was carried out by applying the framework to a wind farm, simulating an application of condition monitoring for bearings through the modelling of its temperature using SCADA data. Three neural networks models were tested: Feedforward Neural Network, Autoregressive Neural Network and Long Short- Term Memory (LSTM). The LSTM model presented the best performance among the tested algorithms, even when compared to other studies, which shows that it can be used for this type of application. The proposed framework is composed by four macro-steps: Selecting application, data preparation, model development and evaluation of results.

BANKING MEMBERS:
Presidente - 2456706 - MARIO ORESTES AGUIRRE GONZALEZ
Interno - 1142787 - JOSE ALFREDO FERREIRA COSTA
Externo à Instituição - ELOI RUFATO JUNIOR - UTFPR
Externo à Instituição - HUMBERTO DIONISIO DE ANDRADE - UFERSA