The Digital Twin for Production Optimization with Emphasis on Compositional Modeling | Kristian Mogensen

Wednesday, October 12, 2022 | 08:00:AM - 08:30:AM CT

Add to Calendar 10/12/2022 08:00 AM 10/12/2022 08:30 AM America/Chicago The Digital Twin for Production Optimization with Emphasis on Compositional Modeling | Kristian Mogensen A digital twin is essentially a digital representation of a physical system such as a well, pump, compressor, or a series of connected items. Sometimes, machine-learning algorithms can assist in analyzing large amounts of data within domains such as preventive maintenance. The value proposition of a digital twin is to have a complete overview of all fluid streams in the production and injection network to enable automation of production capacity planning subject to current and future constraints. The digital solution must be versatile, maintainable, accurate, and with a quick turnaround time to address dynamic changes in market demand as well as the supply side down to the individual wells. Integrated asset models (IAM) have been around for the past two decades or so. These models have become more sophisticated but also require more effort to maintain. The Digital Oil Field (DOF) orchestrates data exchange between different IT systems to feed a calibrated IAM model. An IAM-DOF system provides an up-to-date overview of all fluid streams to maximize value creation subject to a number of constraints. The digital solution must be versatile, maintainable, accurate, and with a quick turnaround time to address dynamic changes in market demand as well as the supply side. So far, digital twins have mainly focused on mimicking small, well-defined systems, whereas IAM models tend to address the bigger picture. Can we take the best from both worlds? Do you need to? And how would you go about developing such a technical solution? Distinguished Lecturer Kristian Mogensen will discuss some of these aspects in the presentation. When you blend different fluid systems together, you need a robust fluid description at well level. Complicating factors may arise, such as compositional variation at reservoir level, gas coning as well as breakthrough of injection gas. Many such technical details must be factored in, without losing sight of the overall goal: to create more value. https://streaming.spe.org/the-digital-twin-for-production-optimization-with-emphasis-on-compositional-modeling

A digital twin is essentially a digital representation of a physical system such as a well, pump, compressor, or a series of connected items. Sometimes, machine-learning algorithms can assist in analyzing large amounts of data within domains such as preventive maintenance.

The value proposition of a digital twin is to have a complete overview of all fluid streams in the production and injection network to enable automation of production capacity planning subject to current and future constraints. The digital solution must be versatile, maintainable, accurate, and with a quick turnaround time to address dynamic changes in market demand as well as the supply side down to the individual wells.

Integrated asset models (IAM) have been around for the past two decades or so. These models have become more sophisticated but also require more effort to maintain. The Digital Oil Field (DOF) orchestrates data exchange between different IT systems to feed a calibrated IAM model. An IAM-DOF system provides an up-to-date overview of all fluid streams to maximize value creation subject to a number of constraints. The digital solution must be versatile, maintainable, accurate, and with a quick turnaround time to address dynamic changes in market demand as well as the supply side.

So far, digital twins have mainly focused on mimicking small, well-defined systems, whereas IAM models tend to address the bigger picture. Can we take the best from both worlds? Do you need to? And how would you go about developing such a technical solution?

Distinguished Lecturer Kristian Mogensen will discuss some of these aspects in the presentation. When you blend different fluid systems together, you need a robust fluid description at well level. Complicating factors may arise, such as compositional variation at reservoir level, gas coning as well as breakthrough of injection gas. Many such technical details must be factored in, without losing sight of the overall goal: to create more value.

Tags: , , , ,

The Digital Twin for Production Optimization with Emphasis on Compositional Modeling | Kristian Mogensen