Banca de DEFESA: FLÁVIO GENTIL DE ARAÚJO FILHO

Uma banca de DEFESA de DOUTORADO foi cadastrada pelo programa.
DISCENTE : FLÁVIO GENTIL DE ARAÚJO FILHO
DATA : 28/07/2017
HORA: 16:00
LOCAL: Auditório do LAUT
TÍTULO:

Optimization of a Fuzzy Locic controller Applied to Oil Wells Equipped with Electrical Submersible Pump

PALAVRAS-CHAVES:

intelligent systems, articial lift, electric submersible pumping, industrial automation, optimization.

PÁGINAS: 115
GRANDE ÁREA: Engenharias
ÁREA: Engenharia Elétrica
SUBÁREA: Eletrônica Industrial, Sistemas e Controles Eletrônicos
ESPECIALIDADE: Controle de Processos Eletrônicos, Retroalimentação
RESUMO:

The aim of this work is to present the development of a set of tuning methods of fuzzy
controllers applied to the nonlinear in oil wells production systems equipped with the electrical
submersible pumping articial lifting method. The ecient control of the studied
industrial plant process will be achieved by the application of articial intelligence techniques,
mainly due to the fact that dierent set-points are required as certain reference
trajectories. The need to obtain expert knowledge in inference and fuzzy logic membership
functions rules is supplied by the technical learning process of neural adaptive control by
inverse model, in which the controller structure and parameters are set automatically,
using for training their own application data. The model neuro-fuzzy obtained is used to
generate control actions in series with the process presenting a suitable dynamic response.
Likewise, techniques based on optimization metaheuristics are employed in order to nd
the optimum or suciently close to optimum solution for the tuning of the fuzzy controller
applied to the petroleum production process. The optimization algorithms of the exible
polyhedron and the simulated annealing were used with dierent types of objective functions
for error system minimization and to select the parameters of the controller tuning
according to the desired performance criteria. The presented results compare the answers
obtained from the system for each one of the techniques employed, evidencing the good
performance and the robustness of the proposed tunings

MEMBROS DA BANCA:
Presidente - 350693 - ANDRE LAURINDO MAITELLI
Interno - 1451883 - FABIO MENEGHETTI UGULINO DE ARAUJO
Externo ao Programa - 1714956 - RUTACIO DE OLIVEIRA COSTA
Externo à Instituição - BENNO WALDEMAR ASSMANN - PETROBRAS
Externo à Instituição - OSCAR GABRIEL FILHO - PETROBRAS