Schulich School of Engineering
Virtual Reality Implementation of Reservoir Models and Integration with Artificial Intelligence
Abstract
dc:description.abstractPromoting forward-looking and proactive thinking is vital in the age of virtual reality (VR) and artificial intelligence (AI). Currently, in the oil and gas industry, traditional ways of thinking and working occupy the minds of most engineers and researchers. With the development of rapid technologies, the impact of virtual reality and artificial intelligence technologies on existing work styles will happen sooner or later. Questions might be raised: How can these technologies integrate or enhance the current workflow? What impact will they have on the work style of petroleum engineers? How would they change their way of thinking and working? This study aims to bring reservoir simulation models from traditional desktop displays to virtual environments and explore further integrations with AI. In order to achieve the transformation, the Unity 3D game engine is selected to develop the VR application. Based on this game platform, the visualization toolkit and mixed reality toolkit are used to implement the visualization of reservoir models and natural interactions in virtual worlds. In order to enhance our understanding of cell-based reservoir data, cross-block properties are studied using the graph theory and visualized upon virtual models. Moreover, different dimensional data can be displayed in the large working space in VR; thus, multi-dimensional and multi-representation visualization features are developed. Further integrations with AI will be explored based on the VR visualization system's digitalized reservoir models. The VR visualization system tests different types of reservoir models and interactions. Interactions and user interfaces are designed and developed with the habits of reservoir engineers in mind. With the assistance of the graph theory, a quick response model is developed. The model can be customized for different purposes, such as conducting and verifying reservoir continuity analysis. In addition, spatial reservoir data and two-dimensional information can be presented in a hybrid visualization mode. Furthermore, two applications, time-series forecasting and refracturing candidate selection, are developed as prototypes for integrating reservoir engineering workflows and AI. It turns out that the combination with AI helps provide a platform to conduct visual data analysis and provides possibilities to conduct data analysis considering both physics and data.
Degree
thesis:*- Name thesis:degree_name
- Doctor of Philosophy (PhD)
- Discipline thesis:degree_discipline
- Engineering – Chemical & Petroleum
- Grantor
- Schulich School of Engineering
- Year dc:date.issued
- 2022
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Hu, Die
- Advisor dc:contributor.advisor
-
- Chen, Zhangxing
- Committee members dc:contributor.committeemember
-
- Chen, Zhangxing
- Chen Shengnan
- Shor Roman J
- Sharlin Ehud
- Feng Yongcun
Subjects
dc:subject × 4Rights
- Language dc:language.iso
- English
Identifiers
dc:identifier.*- Dc Identifier Citation
- Hu, D. (2022). Virtual Reality Implementation of Reservoir Models and Integration with Artificial Intelligence (Doctoral thesis). University of Calgary, Calgary, Canada). Retrieved from https://prism.ucalgary.ca .