sCO2 Turbine and Injection Well CO2 Allocation Using Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The oil and gas industry faces challenges in reducing greenhouse gas emissions from operations such as well activities, with existing methods like recycling waste gases and improved leak detection being insufficient to effectively manage carbon dioxide emissions from supercritical carbon dioxide power turbines and injection wells.
Innovation Solution
A method utilizing a machine-learning model, including a neural network, to predict production and carbon emission data, which adjusts carbon dioxide distribution to achieve a predetermined production rate and carbon footprint by optimizing reservoir management and turbine operations within a carbon dioxide management network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If carbon dioxide is injected into reservoirs for enhanced oil recovery, then hydrocarbon production is improved, but carbon emissions increase
Solution Approach 1:
The system captures carbon emissions that would otherwise be harmful and converts them into a useful resource for enhanced oil recovery operations. By injecting captured CO2 into reservoirs, the system simultaneously reduces emissions and maintains production levels, transforming a waste product into a valuable injection fluid.
Solution Approach 2:
Instead of discarding carbon emissions into the atmosphere, the system recovers and reuses them in CO2 injection operations. The captured CO2 is transported and injected into reservoirs to maintain pressure and enhance hydrocarbon recovery, thereby recovering value from what would otherwise be wasted emissions.
2Object-generated harmful factors
If machine learning models are used to optimize CO2 distribution, then carbon footprint is reduced, but system complexity increases
Solution Approach 1:
The system employs machine learning models that continuously receive feedback from sensors monitoring CO2 injection rates, reservoir pressure, and production outputs. This feedback loop enables the system to automatically adjust CO2 distribution strategies to minimize carbon footprint while maintaining production targets, managing complexity through data-driven automation.
Solution Approach 2:
The machine learning system autonomously optimizes CO2 distribution without requiring constant human intervention. The models self-adjust injection parameters based on real-time data, automatically finding the optimal balance between emission reduction and production maintenance, thereby managing system complexity through self-service optimization.
3Productivity
If real-time adjustments are made to CO2 injection rates, then production rate is maintained, but control system complexity increases
Solution Approach 1:
The control system dynamically adjusts CO2 injection rates in real-time based on changing reservoir conditions and production requirements. Rather than using fixed injection rates, the system continuously adapts parameters such as injection pressure and flow rate to maintain optimal production levels, managing complexity through dynamic rather than static control.
Solution Approach 2:
Real-time sensors monitor production rates and reservoir conditions, providing feedback to the control system. This feedback enables automatic adjustments to CO2 injection rates that maintain production targets while minimizing emissions, managing control complexity through automated closed-loop control rather than manual intervention.
Data Source
AI summary
A method may include obtaining reservoir data for a geological region of interest. The method may further include obtaining turbine data regarding a supercritical carbon dioxide power (sCO2) turbine. The method may further include obtaining carbon emission data for a well coupled to the geological region of interest. The method may further include determining predicted production data and predicted carbon emission data using a machine-learning model, the reservoir data, the turbine data, and the carbon emission data. The method may further include transmitting a command to a control system based on the predicted production data and the predicted carbon emission data. The command may adjusts an amount of carbon dioxide that is distributed to an injection well and the sCO2 turbine. The command achieves a predetermined production rate at the well and a predetermined carbon footprint.