Reservoir Temperature Field Partitioning via Hydrodynamic Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for partitioning reservoir temperature fields are subjective and lack objectivity, leading to inaccurate and ineffective partitioning due to reliance on critical temperature gradients determined by experience, which varies between reservoirs and over time.
Innovation Solution
A method and apparatus that utilize topographic data to determine spatial regions, perform discretization, and apply clustering processing based on water temperature-hydrodynamic feature data to objectively partition the reservoir into regions, reducing subjective interference and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If critical temperature gradient method is used for partitioning, then partitioning process is simple, but partitioning accuracy and objectivity deteriorate due to subjective experience
Solution Approach 1:
The patent replaces the manual, experience-based critical temperature gradient method with an automated clustering algorithm that processes water temperature and hydrodynamic feature data. The clustering unit automatically partitions the reservoir into temperature field regions without requiring subjective judgment on critical gradient values, thereby substituting mechanical/manual operations with an automated computational system that improves both accuracy and objectivity.
2Productivity
If experience-based critical temperature gradient is used, then partitioning can be performed quickly, but reliability and objectivity worsen due to variability between reservoirs and over time
Solution Approach 1:
The patent implements a self-service system where the clustering algorithm automatically determines the optimal partitioning of temperature field regions based on the input data characteristics. The system self-adjusts to different reservoir conditions and time periods without requiring external expert intervention or recalibration of critical gradient thresholds, thereby maintaining high reliability and objectivity across varying conditions while preserving fast processing speed.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The present disclosure provides a method and apparatus for partitioning a reservoir temperature field, a computer device, and a medium. The method for partitioning the reservoir temperature field includes: carrying out discretization on a spatial region of a reservoir to obtain a first number of reservoir region spatial discrete points; based on water temperature-hydrodynamic features of each reservoir region spatial discrete point at different moments and an actual spatial position of each reservoir region spatial discrete point, carrying out clustering processing on the first number of reservoir region spatial discrete points to obtain a second number of reservoir temperature field regions and a spatial distribution of each reservoir temperature field region; and according to a corresponding actual spatial position of each reservoir temperature field region, determining a region type corresponding to each reservoir temperature field region. According to the present disclosure, an obtained reservoir temperature field partitioning result does not need to be depended on experimental parameters, reservoir temperature field partitioning based on a specified time interval has higher objectivity and stability, and the powerful technical support can be provided for actual operation and scheduling of the reservoir.