Machine Learning SIG Meeting – April (Virtual Event)

PESGB members can register free of charge via the membership platform

If you are interested in becoming a member of PESGB please visit: https://www.ges-gb.org.uk/membership/

Speaker: Konrad Wojnar – Resoptima
Topic: Assimilation of 4D seismic data for probabilistic modelling using ensemble-based methods

Abstract
A standardized approach to application of 4D seismic data to reservoir characterization and history matching is yet to be adopted by the industry. In this talk I will present an ensemble-based workflow to systematically screen, analyze and history match production and 4D seismic data. I will focus on why ensemble-based methods are better suited for data with high dimensionality and why they are less prone to overfitting than traditional machine-learning data conditioning algorithms. Moreover, I will discuss how the workflow can be used to explore different modelling scenarios and thus improve reservoir management.

Quantitative 4D Seismic Assisted History Matching Using Ensemble-based Methods On The Vilje Field, 2020, Konrad Wojnar et al., SPE Annual Technical Conference and Exhibition

Konrad Wojnar – Resoptima

Biography
Konrad Wojnar is currently a senior reservoir engineer at Resoptima. His focus is on helping clients establish probabilistic modelling workflows for improved reservoir management. For the past four years his main role in Resoptima has been hands-on involvement in technology implementation projects in several assets in Europe and Africa.