Reservoir Modeling Platform for Clustered Well Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing reservoir modeling technologies face challenges in efficiently handling large datasets from numerous wells, requiring manual data cleaning and laborious visualization, and struggle with accurately predicting infill and sidetrack well locations due to computational complexity and limited scalability.
Innovation Solution
An integrated software system employing automation, machine learning, and artificial intelligence for reservoir modeling, which includes data analytics to clean datasets, perform history matching, and automatically identify infill and sidetrack wells, while providing real-time visualization and permeability calibration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual data cleaning and laborious visualization are used, then measurement precision can be maintained, but productivity and time efficiency deteriorate
Solution Approach 1:
The system performs automated data cleaning by detecting and removing statistical outliers without manual intervention. The automation engine processes data records, identifies anomalies based on predefined criteria, and cleans the dataset automatically, allowing the system to serve itself rather than requiring manual data processing by experts.
Solution Approach 2:
Manual data cleaning operations are replaced by computational algorithms that automatically detect and remove statistical outliers. The system uses programmed logic and statistical methods to perform what would otherwise require manual inspection and processing, substituting human mechanical work with automated computational processes.
2Measurement precision
If manual visualization and analysis are performed, then measurement precision is maintained, but loss of time increases
Solution Approach 1:
The system continuously updates the 3D visualization as data processing and simulation progress, rather than requiring discrete manual updates. The automation engine maintains continuous computation and rendering operations, ensuring that the visualization is always current without requiring periodic manual intervention to refresh or update the display.
Solution Approach 2:
Manual visualization operations are replaced by automated computational rendering that continuously generates and updates 3D representations. The system uses computer-generated graphics and automated data processing to create visualizations that would otherwise require manual creation and updating by analysts.
3Adaptability or versatility
If computational complexity is reduced for scalability, then ease of operation improves, but measurement precision and prediction accuracy deteriorate
Solution Approach 1:
The system divides the reservoir into multiple clusters based on geological and production characteristics. This segmentation allows the complex reservoir to be modeled in manageable sections, each with its own simulation parameters, enabling the system to scale to large reservoirs while maintaining computational tractability and prediction accuracy through localized analysis.
Solution Approach 2:
The system dynamically adjusts simulation parameters and computational complexity based on the specific characteristics of each reservoir cluster. By changing parameters such as simulation resolution, computational methods, and data processing depth according to local conditions, the system maintains high prediction accuracy while adapting to varying scales and complexities of different reservoirs.
Data Source
AI summary
Implementations provide a method that includes: accessing data comprising records of measurements from a plurality of wells of a reservoir over a period of time; removing statistical outliers from the records to generate clean records, wherein the statistical outliers represent a probability lower than a threshold; grouping, based on the clean records, the plurality of wells into a set of clusters, each cluster comprising one or more wells whose corresponding records exhibit a shared trend over at least a portion of the period of time; conducting, for each cluster, a history matching simulation using a corresponding model, wherein the corresponding model is calibrated; launching, based on results of the history matching simulation, a prediction simulation to identify at least one of an infill well and a sidetrack well within each cluster; and generating an integrated visualization for results of the prediction simulation as the prediction simulation advances.