SAGD Control System Reducing Model Uncertainty for Production Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Steam-assisted gravity drainage (SAGD) oil well systems face challenges in optimizing emulsion production due to high uncertainty in predictive models, which are affected by complex geology and limited data, leading to biased predictions and inefficient steam distribution.
Innovation Solution
A method is developed to generate and train models using historical time series data from sensors, determining control variables that reduce model uncertainty while meeting objectives and constraints, thereby optimizing SAGD system controls to enhance emulsion production efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If predictive models are used for SAGD system control, then production optimization is enabled, but model uncertainty leads to biased predictions and inefficient steam distribution
Solution Approach 1:
The system implements feedback by continuously monitoring sensor data from the SAGD wells and using it to update the predictive model. The model predictions are compared against actual sensor measurements, and the discrepancies are used to adjust model parameters and reduce uncertainty, creating a closed-loop control system that improves prediction reliability over time.
Solution Approach 2:
The system dynamically adjusts model parameters based on sensor data and operational conditions. By changing parameters such as steam injection rates, well pressure, and temperature settings, the system explores different operational states to gather more data and reduce model uncertainty, thereby improving prediction accuracy without sacrificing productivity.
2Productivity
If steam injection is increased to improve bitumen recovery, then emulsion production increases, but water-to-bitumen ratio and steam requirements worsen
Solution Approach 1:
The system optimizes steam injection parameters by dynamically adjusting injection rates, pressure, and temperature based on real-time sensor data and model predictions. This allows the system to achieve high bitumen recovery while minimizing the water-to-bitumen ratio by finding the optimal steam injection parameters that maximize efficiency.
Solution Approach 2:
The control system transitions from static steam injection schedules to dynamic adjustment of steam parameters. The system continuously adapts steam injection rates and distribution based on changing reservoir conditions, sensor feedback, and model predictions, enabling efficient bitumen recovery with reduced water and steam requirements over time.
3Reliability
If model exploration is prioritized to reduce uncertainty, then long-term efficiency improves, but short-term production objectives may be compromised
Solution Approach 1:
The system dynamically balances exploration and exploitation by adjusting the weight given to model uncertainty reduction versus production optimization based on the current state. When model uncertainty is high, the system prioritizes data collection and model improvement; when uncertainty is reduced, it shifts focus to maximizing production, creating a time-varying control strategy.
Solution Approach 2:
The system implements periodic cycles of model exploration and production optimization. During exploration phases, the system gathers data and reduces uncertainty; during exploitation phases, it maximizes production using the refined model. This periodic alternation ensures long-term efficiency improvement while maintaining acceptable short-term production levels.
Data Source
AI summary
A method for increasing efficiency in emulsion production for a steam-assisted gravity drainage (SAGD) oil well system includes generating a model of the SAGD oil well system and training the model of the SAGD oil well system utilizing historical time series data relating to one or more SAGD oil wells at one or more SAGD production sites of the SAGD oil well system. The historical time series data is obtained from a plurality of sensors in the SAGD oil well system. The method also includes determining an uncertainty of the model as a function of a control space of the model and utilizing the model to determine values for control variables associated with one or more of the SAGD production sites of the SAGD oil well system which reduce the model uncertainty while meeting one or more objectives subject to one or more constraints. The method further includes adjusting a set of controls of the SAGD oil well system based on the determined values for the control variables.