Flow Control Opening Layout Using Clustered Reservoir Settings
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for designing flow control devices in oilfields are computationally expensive and inefficient, particularly when using greedy algorithms with reservoir simulators to determine discrete configurable positions, which hinders optimal hydrocarbon extraction.
Innovation Solution
A computer-implemented method that uses cluster analysis to determine a set of discrete flow control device configurable positions based on continuous space settings, increasing computational efficiency and guiding the optimizer towards an optimized solution by re-inputting cluster-determined sets into the reservoir simulator.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If greedy algorithms with reservoir simulators are used to determine discrete configurable positions of flow control devices, then the design accuracy is improved, but the computational cost and time increase significantly
Solution Approach 1:
The patent performs preliminary cluster analysis on continuous space settings to identify discrete configurable positions before running the reservoir simulator. This pre-processing step groups similar continuous settings into clusters, determining representative discrete positions in advance, which guides the subsequent optimization process and reduces the number of simulator runs needed.
Solution Approach 2:
The patent introduces cluster analysis as an intermediary step between the continuous space settings and the discrete configurable positions determination. This intermediary process transforms the continuous optimization problem into a discrete one by identifying representative positions through clustering, thereby reducing the computational burden on the reservoir simulator while maintaining design accuracy.
2Adaptability or versatility
If continuous space settings are used for flow control device optimization, then the solution space is comprehensive, but the computational complexity increases
Solution Approach 1:
The patent segments the continuous solution space into discrete clusters through cluster analysis. By dividing the continuous space of possible flow control device settings into distinct groups, it identifies representative discrete positions for each cluster. This segmentation maintains comprehensive coverage of the solution space while reducing computational complexity by working with discrete rather than continuous variables.
Solution Approach 2:
The patent changes the parameter representation from continuous space coordinates to discrete cluster labels. Instead of optimizing over continuous flow control settings, the method transforms the problem into selecting from discrete configurable positions identified through clustering. This parameter transformation reduces computational complexity while preserving the essential characteristics of the continuous optimization problem.
Data Source
Figure 1
Figure 2
Figure 3.1~3.2
AI summary
Defining flow control device configurable positions include executing a reservoir simulator on a reservoir model to obtain a collection of flow control device settings defined in continuous space. For each number of multiple numbers of clusters, a cluster analysis is individually performed on the collection to obtain a set of flow control device configurable positions. The set includes the number of flow control device configurable positions and its corresponding inflow area or diameter. Performing the cluster analysis across the numbers generates multiple sets for the multiple numbers of clusters. The sets of flow control device configurable positions are compared to obtain a selected set of flow control device configurable positions, which is presented in a completion design.