Gas Lift Injection Range Prediction for Oil Well Production
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for optimizing gas lift injection rates in oil wells, such as bottom hole pressure surveys and trial-and-error approaches, result in production losses, delays, and non-optimized values due to the need for well interventions and reliance on historical data.
Innovation Solution
Utilizing machine-learning models trained on synthetic datasets generated from well, reservoir, and historical production data to predict the applicability of gas lift injection, an optimal range of gas lift values, and an optimal liquid production rate, eliminating the need for well interventions and biases in traditional methods.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If bottom hole pressure survey is conducted to determine optimal gas lift injection rate, then measurement precision is improved, but loss of time increases and productivity decreases
Solution Approach 1:
The system performs preliminary actions by collecting well data, reservoir data, and historical production data before optimization is needed. Machine learning models are trained in advance on synthetic datasets generated from this data, so when optimization is required, the system can quickly query pre-trained models without conducting time-consuming well interventions.
Solution Approach 2:
The patent creates synthetic datasets that copy and simulate real well production scenarios. These synthetic datasets are generated from well data, reservoir data, and historical production data, allowing the machine learning models to learn from numerous simulated cases without requiring actual well interventions for each optimization case.
2Reliability
If bottom hole pressure survey is conducted to determine optimal gas lift injection rate, then reliability of optimization is improved, but productivity decreases
Solution Approach 1:
The system incorporates feedback loops where production test history data is continuously fed into the machine learning models. The models learn from historical performance data and adjust predictions based on actual well responses to gas lift injection, improving reliability while allowing continuous production without shutdowns.
Solution Approach 2:
The patent replaces the mechanical well intervention system (slickline units, pressure gauges, shut-in procedures) with an information-based system using machine learning models. These models process well data, reservoir data, and historical production data to predict optimal gas lift injection rates without requiring physical well interventions.
3Ease of operation
If trial-and-error approach is used to optimize gas lift injection rate, then ease of operation is improved, but loss of time increases and manufacturing precision worsens
Solution Approach 1:
The system performs self-service by automatically determining optimal gas lift injection rates using machine learning models. The models take well data, reservoir data, and historical production data as inputs and directly output optimized injection rates, eliminating the need for manual trial-and-error adjustments by engineers.
Solution Approach 2:
The patent utilizes parameter changes in the machine learning models to transition from imprecise trial-and-error methods to precise predictions. The models learn optimal parameter relationships from training data and can predict optimal gas lift injection rates with high precision by changing input parameters such as well depth, reservoir pressure, and historical production rates.
4Measurement precision
If multiple BHP surveys are conducted for multiple wells, then measurement precision is improved for each well, but loss of time increases substantially
Solution Approach 1:
The machine learning models serve multiple functions across different wells. A single trained model can predict optimal gas lift injection rates for any well that provides the required input data (well data, reservoir data, historical production data), eliminating the need to conduct separate BHP surveys for each well while maintaining consistent prediction quality.
Solution Approach 2:
The system uses synthetic datasets that copy real well scenarios to train universal models. These models learn from numerous simulated well cases and can then be applied to actual wells without requiring actual surveys, allowing the same modeling approach to serve multiple wells simultaneously.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables efficient and reliable determination of optimal gas lift injection rates and liquid production rates, reducing production losses and delays, and improving oil production efficiency by automating the optimization process.
Implementation Method 1
Gas lift is an artificial lift method that uses gas to facilitate the lifting of liquids from wellbore to the surface. The gas, which is often natural gas, is pressurized at the surface then injected into the wellbore. The injected gas reduces the density of the fluid in the tubing or inner casing, which lowers a flowing bottom hole pressure of the wellbore and results in higher liquid production.
Data Source
AI summary
Embodiments relate to acquiring well, reservoir and production data, synthesizing training and test data, and then constructing, training, and utilizing machine-learning models to: (i) predict whether or not gas lift should be applied to facilitate the production of subsurface fluids from an oil well, (ii) predict an optimal range of gas lift values to be used in the production of fluids from the oil well, and (iii) predict an optimal liquid production rate for the oil well when the gas lift value is within the predicted optimal range. Unlike traditional approaches, the disclosed embodiments do not require the use of well interventions, which eliminates production losses, delays, and costs. The disclosed embodiments also avoid the delays, biases, and non-optimized values associated with existing trial-and-error-based approaches to gas lift injection optimization. Disclosed embodiments enable the efficient and reliable determination of the optimal range of gas lift values and the optimal liquid production rate.