Inverse Wellbore Flow Modeling for Reservoir Pressure Inference
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for estimating near wellbore reservoir pressures and downhole parameters are limited by low frequency data and do not account for changes in artificial lift setpoints, leading to reduced accuracy and inefficiency in well performance optimization.
Innovation Solution
An automated machine learning approach using inverse modeling and neural networks for high-resolution simulation, incorporating live sensor data and adaptive learning to estimate near wellbore pressures and optimize well performance by identifying stable periods, simulating states, matching simulations with field data, and updating models through drift modeling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If automated inverse wellbore flow modeling is implemented, then measurement precision of near wellbore reservoir pressure is improved, but device complexity increases
Solution Approach 1:
The patent introduces an automated inverse wellbore flow modeling system that acts as an intermediary between available production data and near wellbore reservoir pressure estimation. This intermediary system uses neural networks and drift modeling to bridge the gap between surface measurements and downhole conditions, achieving high-frequency pressure estimates without direct downhole gauge installation.
Solution Approach 2:
The patent replaces physical downhole gauges and probe tests with a computational modeling approach. Instead of using mechanical sensing devices in the wellbore, the system substitutes these with automated inverse modeling that processes production data through neural networks to infer downhole pressure conditions, thereby avoiding the complexity of physical downhole instrumentation.
2Productivity
If high-frequency production data is used for inverse modeling, then productivity is improved, but loss of information increases due to artificial lift setpoint changes
Solution Approach 1:
The patent extracts and isolates the effect of artificial lift setpoint changes from the production data before performing inverse modeling. By separating the known setpoint influence from the residual production behavior, the system can focus the inverse modeling on reservoir-driven signals, thereby reducing information loss and improving the accuracy of near wellbore pressure estimates.
Solution Approach 2:
The patent performs preliminary processing of production data to account for artificial lift setpoint changes before applying inverse modeling. By pre-adjusting the data to remove the known effects of setpoint variations, the system prepares cleaner input data for the inverse modeling process, reducing information loss and improving productivity of the analysis.
3Reliability
If reservoir simulation based inverse modeling is used, then reliability is improved, but loss of time increases due to low-frequency data requirements
Solution Approach 1:
The patent transitions from static, low-frequency reservoir simulation to a dynamic automated inverse modeling approach that processes production data at high frequency. The system adapts to changing production conditions in real-time, updating near wellbore pressure estimates dynamically rather than relying on periodic simulation runs, thereby reducing time loss while maintaining reliability.
Solution Approach 2:
The patent changes the temporal parameter of data processing from low-frequency periodic simulation to high-frequency continuous modeling. By altering the time scale at which inverse modeling is performed, the system achieves both improved time resolution and maintained reliability, allowing frequent updates of reservoir pressure estimates without the computational burden of full reservoir simulations at each interval.
Data Source
AI summary
A method to estimate the likely downhole conditions in the wellbore and reservoir by Inverse modeling well flow simulation history matched with field sensor data. The invention presents a method for automating sensor data processing through cleaning, transformation, and identification of stable states. This process is crucial for the selection of data to be simulated and matched without human review. The matched simulations are subjected to a state-space model in order to assign a probability to a given unknown state. This probability is updated at each time step. As the well undergoes transition over time including decline, the drift of the likely state of operation is orchestrated to allow physically constrained movement to a proximate space. Based on the extent of repetition and overlap between similar states as they transition over several time steps, the confidence of the inverse model increases, thus narrowing down the likely domain and trajectory of operation and boosting the probability of this narrowed zone. The knowledge of downhole and near wellbore reservoir zone is essential for better modeling, understanding of the wells and decision making in the oilfield. This knowledge may be obtained through well testing but involves physical intervention that can involve expense and production loss. It is also less common to have such well tests being performed at a daily, weekly or even monthly basis so timely information is generally not available. This invention provides a mechanism to have a live update of such information without any physical intervention.