Banca de QUALIFICAÇÃO: JOSENÍLSON GOMES DE ARAÚJO

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : JOSENÍLSON GOMES DE ARAÚJO
DATE: 04/06/2021
TIME: 17:00
LOCAL: Banca remota via Google Meet. Link: meet.google.com/jna-vkwo-ssg
TITLE:

APPLICATION OF DEEP LEARNING MODELS IN OIL FLOW RATES DATA IN ELECTRICAL SUBMERSIBLE PUMP SYSTEM

KEY WORDS:

fluid flowrate measurement, virtual flow metering (VFM), electric submersible pumps (ESP), deep learning, LSTM neural networks, Industry 4.0.

PAGES: 49
BIG AREA: Engenharias
AREA: Engenharia de Energia
SUMMARY:

The fluid flow measurement is a fundamental activity for the oil and gas industry. The correct produced volumes mensuration provides a good reservoirs management, reducing production losses, guiding plans of the production system optimization and of the lift and production flow methods. Despite its great importance, the produced fluids measurement, in general, is still carried out through physical multiphase flow meters. This methodology ends up leading to a limitation of physical resources and requires relatively long test time to provide accurate results. In this context, the use of flow estimation techniques in real time using Virtual Flow Metering (VFM) has shown to be a promising field due to the provided results precision either their low cost implementation. These techniques have been benefited by the large data volume collected through sensors and transmitters in the wells, in other words, the development of VFM methods is related to the digital transformation that the industry is currently experiencing, called the fourth industrial revolution or yet Industry 4.0. Combining technological advances and the great importance of fluid measurement for the oil industry, the study aims to develop a model for the flowrate of liquid through neural networks Deep Learning specifically the Long Short-Term Memory (LSTM) type. Has been used data collected from two offshore wells use electric submersible pumps (ESP) in the state of Rio Grande do Norte. LSTM forecasts compare favorably with the results of hydrodynamical modeling. The provided results can be useful to accurately estimate the flowrate behavior in real time and to forecast the flowrate for a sequence of future time instants, supporting better production management. It is expected that the results obtained with the LSTM neural networks can be integrated with other technologies of Industry 4.0 and contribute to the digital transformation of the oil and gas industry.

BANKING MEMBERS:
Presidente - 6350734 - CARLA WILZA SOUZA DE PAULA MAITELLI
Interno - 347628 - ADRIAO DUARTE DORIA NETO
Externo à Instituição - FABIO SOARES DE LIMA - PETROBRAS