Hydrocarbon Well Flow Instability Detection Using Surrogate Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Slugging, a type of unstable flow, occurs in gas lift wells, leading to decreased production rates, facility upsets, and potential shut-down risks, and existing detection and mitigation techniques are complex and require data-driven surrogate models for accurate diagnosis and improvement.
Innovation Solution
A method involving real-time production data analysis to generate and calibrate numerical models of transient and thermal multiphase flow, performing parametric studies to determine instability types and optimal operating conditions, and providing advisory recommendations to improve well stability and performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If accurate slugging detection and mitigation techniques are used, then production rate and system reliability are improved, but device complexity and difficulty of detecting and measuring increase due to requiring data-driven surrogate models and careful history matching procedures
Solution Approach 1:
The patent creates a simplified copy (surrogate model) of the complex well system that replicates slugging behavior patterns. This surrogate model captures essential dynamics without requiring full complexity of the original system, enabling detection and mitigation while reducing computational and operational complexity
Solution Approach 2:
The patent replaces complex mechanical analysis and manual history matching procedures with data-driven computational models. By substituting traditional mechanical diagnostic methods with automated surrogate models that process production data, the system achieves accurate slugging detection while reducing operational complexity
2Productivity
If accurate slugging detection and mitigation techniques are used, then production rate and system reliability are improved, but the difficulty of detecting and measuring slugging patterns increases due to requiring careful history matching procedures
Solution Approach 1:
The surrogate model creates a simplified representation that copies essential slugging patterns from historical data, making detection easier by focusing on key diagnostic features rather than requiring comprehensive analysis of all production parameters
Solution Approach 2:
The system performs self-diagnosis by automatically analyzing its own production data through the surrogate model. The model autonomously identifies slugging patterns and provides mitigation recommendations without requiring external expert intervention, reducing the difficulty of detection and measurement
Data Source
AI summary
A method of detecting and mitigating flow instabilities, such as slugging, in hydrocarbon production wells. Real-time production data pertaining to each well is are retrieved. Using the production data, patterns of flow instability are identified therein. A numerical model of transient and thermal multiphase flow in each well is generated. Well test data is are retrieved from a database. The numerical model is calibrated using the well test data. Using the calibrated numerical model, a parametric study is performed to determine how input parameters affect at least one of stability and performance of the wells. Results of the parametric study are queried to determine a type of flow instability and to determine operating conditions to improve performance of the wells. An advisory is provided to a user to change operating conditions of one or more of the wells, to improve stability and/or performance of one or more of the wells.