Banca de DEFESA: KATERINE DE JESUS RINCON PEREZ

Uma banca de DEFESA de DOUTORADO foi cadastrada pelo programa.
STUDENT : KATERINE DE JESUS RINCON PEREZ
DATE: 26/08/2025
TIME: 14:00
LOCAL: meet.google.com/ezg-agtf-euw
TITLE:

Deep Learning Applications in Geophysics: Data Matching and Inversion

KEY WORDS:

Seismic time-lapse, OBN, Deep Learning, non-repeatability effects, inversion, FWI.

PAGES: 107
BIG AREA: Ciências Exatas e da Terra
AREA: Ciência da Computação
SUMMARY:

This work applies Deep Learning (DL) methodologies to solve geophysical problems, with a focus on Time lapse seismic, an essential technique for monitoring changes in reservoirs and supporting strategic management, using synthetic pre-salt seismic data acquired with Ocean Bottom Nodes (OBN). The first approach focused on mitigating one of the main limitations of TL seismic: data non-repeatability between surveys, caused by various environmental factors and equipment positioning variations, which can mask real reservoir changes. To address this challenge, we proposed Conditional Generative Adversarial Networks (cGANs). This methodology was developed to correct non-repeatability effects, improving data quality before applying Double Difference Full Waveform Inversion (DDFWI). As a second application, aiming to estimate velocity variations and as a technique parallel to inversion, we built a synthetic dataset representing scenarios with different velocity anomalies. A Convolutional Neural Network (CNN) was trained using synthetic baseline and monitor seismograms generated from a pre-salt velocity model. The network input was the time-lapse difference between baseline and monitor data, and the output was the corresponding velocity model. Results show that the CNN achieved accurate inversion performance with significantly higher computational efficiency. Both scenarios were trained with 80% of the data and tested with the remaining 20%. The predictions were evaluated using quantitative metrics such as the mean squared error (MSE), the normalized root mean squared error (NRMS), and the structural similarity index (SSIM), indicating a good inference capacity of the model.

COMMITTEE MEMBERS:
Externo à Instituição - BRUNO SOUZA CARMO
Externo ao Programa - 1379465 - GILBERTO CORSO - UFRNExterno à Instituição - HERMES SENGER
Externo ao Programa - 2492756 - JOAO MEDEIROS DE ARAUJO - UFRNInterno - 1153006 - LUIZ AFFONSO HENDERSON GUEDES DE OLIVEIRA
Presidente - 1673543 - SAMUEL XAVIER DE SOUZA