Well Shut-in Identification via Pressure Data Pattern Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for identifying shut-ins of a well are inaccurate, leading to false positives and negatively impacting decision-making, surveillance, and regulatory compliance in subterranean operations.
Innovation Solution
A method and system for automatically identifying shut-ins of a well by obtaining raw pressure values, applying filters to generate filtered pressure values, identifying initial sequences with monotonically trending values, splicing adjacent sequences, adjusting sequence start points based on pressure derivatives, and applying parametric functions and pattern recognition to refine the identification of shut-ins.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple filtering and analysis steps are applied to pressure data, then measurement precision of shut-in identification is improved, but device complexity increases
Solution Approach 1:
The pressure data analysis is divided into multiple sequential processing stages: initial filtering to remove noise, sequence identification to detect monotonic trends, splicing to combine adjacent sequences, derivative-based start point adjustment for precision, parametric function fitting to model pressure behavior, and pattern recognition to classify shut-in events. Each stage processes the output of the previous stage, progressively refining the identification accuracy while maintaining a structured approach to complexity management.
2Productivity
If automated pattern recognition is applied to pressure data, then productivity of shut-in identification is improved, but measurement precision may deteriorate due to false positives
Solution Approach 1:
The system incorporates multiple feedback mechanisms to reduce false positives while maintaining automated processing. Pattern recognition algorithms compare detected sequences against established shut-in patterns, and parametric function fitting evaluates whether pressure data conforms to expected shut-in behavior. The derivative-based start point adjustment provides additional verification by ensuring pressure changes align with theoretical shut-in dynamics. These feedback loops filter out false positives that would otherwise pass through automated processing.
Solution Approach 2:
The system transforms raw pressure data through multiple parameter transformations: computing pressure derivatives to identify transition points, fitting parametric functions to model pressure evolution, and extracting characteristic features for pattern matching. These parameter changes convert raw data into forms that are more amenable to automated recognition while preserving the essential shut-in characteristics, enabling high-speed processing without sacrificing accuracy.
Data Source
AI summary
A method for identifying shut-ins of a well may include obtaining raw pressure values captured over a period of time, where each of the raw pressure values indicates a pressure at a bottom hole of the well; applying filters to the raw pressure values to generate a plurality of filtered pressure values; identifying initial sequences having monotonically trending values among the plurality of filtered pressure values; splicing adjacent initial sequences; adjusting the start point of at least one of the initial sequences based on derivatives of the filtered pressure values; applying a parametric function to the initial sequences to generate fitted sequences; and applying a pattern recognition to the fitted sequences to generate final sequences.