Disputas: Ali Moradi

Ali Moradi skal forsvare avhandlingen sin for graden philosophiae doctor (ph.d.) ved Universitetet i Sørøst-Norge.

Han har fulgt doktorgradsprogrammet prosess, energi og automatiseringsteknikk ved Fakultetet for teknologi, naturvitskap og maritime fag.

Alle interesserte ønskes velkommen til prøveforelesning og disputas.

• Her kan du lese hele avhandlingen

Sammendrag

Achieving cost-effective and energy-efficient oil recovery depends heavily on optimized well completion design and suitable control of well performance around the right production targets. Advanced wells equipped with flow control devices significantly influence multiphase flow behavior within the well and its adjacent reservoir. Determining appropriate production targets requires a dynamic model that considers the long-term behavior of the reservoir and production system, which are subject to significant uncertainty. To address these challenges, this PhD thesis presents an integrated dynamic modeling approach to simulate and optimize long-term oil production from advanced wells while incorporating the inherent uncertainties in reservoir behavior.

The thesis initially develops fully integrated dynamic well-reservoir models using mechanistic approaches implemented in dynamic multiphase flow simulators. These models accurately capture the performance of advanced wells under a range of operating conditions, including transient phenomena such as coning and complex multiphase interactions. However, their high computational demand limits their practicality for tasks like uncertainty quantification and optimization, which require large ensembles of simulations.

To address this limitation, the functionality of physics-based models such as the MultiSegment Well (MSW) model, as well as data-driven proxy models based on Artificial Neural Networks (ANNs) were explored. These models significantly reduce simulation time, enabling efficient uncertainty quantification with an acceptable trade-off in accuracy for many practical scenarios.
The thesis also proposes a comprehensive workflow for side-by-side modeling and analysis of different advanced well completion strategies with various flow control device configurations. This workflow supports optimization studies aimed at improving the design of advanced multilateral wells and enhancing long-term oil recovery. Furthermore, new techniques were developed to extend the capabilities of open-source tools for modeling complex integrated well-reservoir systems, overcoming the limitations of commercial alternatives.

The findings of this research highlight the importance of integrating advanced well-reservoir modeling, uncertainty quantification, and optimization techniques to maximize oil production.