Reservoir Staging Index for Automated Stage Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional stage selection methods in wells often result in poor quality stages that are difficult to inject fluid into or fracture, leading to increased costs and operational inefficiencies, and are prone to human error due to high dependency on human interaction and inconsistent rock type injectivity models.
Innovation Solution
A computer-implemented method for reservoir stage selection that determines rock type and transmissible sublayers using petrophysical data, applies conditional formatting to identify potential stages, and reduces human interaction by integrating petrophysical and geomechanical characterization to optimize stage selection, incorporating rock type injectivity models for improved injectivity and stimulation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional stage selection methods are used, then human interaction is required for stage selection, but this leads to human error and inconsistent selections
Solution Approach 1:
The patent replaces the mechanical human decision-making process with an automated computer system that uses petrophysical data and machine learning models to select stimulation stages. The system automatically processes well logs, applies rock type injectivity models, and generates stage selections without human intervention, eliminating inconsistencies caused by different engineers' subjective judgments.
Solution Approach 2:
The system performs self-service by automatically evaluating petrophysical parameters, applying cut-off criteria, and generating stage selections independently. The machine learning model continuously learns from offset well performance data, enabling the system to improve its selection accuracy autonomously without requiring human retraining or intervention.
2Productivity
If manual stage selection is performed, then engineers can optimize based on experience, but this increases operational time and reduces efficiency
Solution Approach 1:
The system performs preliminary actions by pre-processing well logs, automatically identifying rock types, and pre-calculating petrophysical parameters before stage selection is needed. The machine learning model is pre-trained on historical offset well data, enabling rapid predictions of injectivity and stimulation success without requiring time-consuming manual analysis during the actual selection process.
Solution Approach 2:
The patent replaces time-consuming manual engineering analysis with automated computer processing that instantly evaluates petrophysical data, applies rock type injectivity models, and generates stage selections in seconds. The system processes multiple well logs and parameters simultaneously, achieving high productivity without sacrificing the optimization that engineers previously provided through experience.
3Reliability
If rock type injectivity models are incorporated, then injectivity and stimulation quality improve, but the system complexity increases
Solution Approach 1:
The patent segments the complex rock type injectivity model into discrete, manageable components. The system divides the reservoir into distinct rock types (e.g., carbonate, shale, sandstone) and applies specific cut-off criteria and petrophysical parameter thresholds to each rock type. This segmentation allows the complex model to be processed systematically by the computer while maintaining high prediction accuracy for each rock type's injectivity characteristics.
Solution Approach 2:
The system manages complexity by changing parameters dynamically based on rock type identification. Different petrophysical parameters and cut-off values are applied depending on the identified rock type (e.g., different porosity and permeability thresholds for carbonate versus shale). The machine learning model adjusts its prediction parameters automatically based on the input data characteristics, maintaining high accuracy without requiring manual adjustment of complex model parameters.
Data Source
AI summary
Systems and methods include a method for determining perforation cluster points of a reservoir. A rock type of a geological formation of a reservoir is determined using petrophysical data. The transmissible sublayers of the reservoir are determined by grouping the petrophysical data into different subset transmissible layers based on cut-off criteria. Net transmissible reservoir footage for the transmissible sublayers are generated based on averages of parameters for minimum and maximum measured depths. Potential stages for the reservoir are determined using a conditional formatting on the transmissible reservoir footage. Potential perforation cluster points selected based on the potential stages for the reservoir are received from input of an engineer.