Banca de QUALIFICAÇÃO: KALINE JULIANA SILVA DO NASCIMENTO

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : KALINE JULIANA SILVA DO NASCIMENTO
DATE: 14/06/2021
TIME: 09:00
LOCAL: Virtual Pelo Google Meet
TITLE:

DEEP Q-NETWORK IN THE DEDECISION PROCESS FOR THE USE OF THE RECOVERY METHOD OF WATER INJECTION IN AN OIL FIELD

KEY WORDS:

Oil field. Intelligent system. Deep Reinforcement Learning. Deep Q-Network. Economic analysis.

PAGES: 60
BIG AREA: Engenharias
AREA: Engenharia de Energia
SUMMARY:

It is necessary, for the best oil production, the constant development of new alternatives for the exploitation of the fields. The need to optimize the factors that are part of this process requires great care in all the recommendations proposed for this purpose. Among the elements that make up the oil exploration are important to highlight: Number of wells, space between them, mesh model, supplementary fluid injection system, among others. This research work aims to present the development and application of an intelligent system based on the Deep Reinforcement Learning technique in oil reservoirs using the water injection method. The simulation was carried out with the mathematical simulator STARS (Steam Thermal ans Advanced Process Reservoir Simulator) of CGM (Computer Modeling Group) with some similar data from homogeneous and semi-synthetic reservoirs found in Northeast Brazil, with dimensions 400mx400mx26m, 23% porosity and permeability between 40 and 400 mD. The algorithm applied was the Deep Q-Network (DQN) which aims to find the optimal policy with the reward of maximizing the Net Present Value (NPV) and the significant increase in the Recovery Factor with the actions of increasing or not the flow of injection of water at the beginning of production over a production horizon estimated at 240 months (20 years). With the results it can be seen that the optimum water injection policy made possible significant increases in the field recovery factor, as well as in the NPV, in addition to obtaining a better profitability, with the costs of water injection, treatment and disposal of produced water, thus increasing the project's feasibility time.

BANKING MEMBERS:
Presidente - 347628 - ADRIAO DUARTE DORIA NETO
Externo ao Programa - 1754344 - MARCOS ALLYSON FELIPE RODRIGUES
Externo à Instituição - THIAGO HENRIQUE FREIRE DE OLIVEIRA
Externo à Instituição - WILSON DA MATA