Wireline Run Duration Modeling via ML and Physics Fusion

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current hydrocarbon recovery operations face challenges in efficiency and cost, particularly due to the need for highly experienced field personnel and the increasing complexity of smaller hydrocarbon deposits.

Innovation Solution

The implementation of a machine learning-based method for modeling run durations in wireline operations, which involves gathering data on wellbores and conveyances, separating run duration into winch and logging times, and using algorithms and models to calculate these durations accurately.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If highly experienced field personnel are used for planning and executing hydrocarbon recovery operations, then operation efficiency and safety are improved, but labor costs increase

Engineering Contradiction:

Improveoperation efficiencyVSAvoidlabor cost

Core Design Contradiction:

ProductivityVSQuantity of substance

Solution Approach 1:

The patent creates a virtual copy of experienced field personnel's knowledge through machine learning models. The system trains AI models on historical data from experienced operators to replicate their planning and decision-making capabilities, allowing the system to perform operations without requiring expensive human expertise for each task.

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent replaces the mechanical system of human operators with an automated machine learning-based system. The ML models process historical operation data, predict optimal workflows, and generate planning recommendations automatically, substituting human cognitive functions with computational algorithms that can scale without additional cost.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Productivity

If conventional planning methods are used for hydrocarbon recovery operations, then simplicity is maintained, but the ability to optimize based on previous project data is limited

Engineering Contradiction:

Improveoptimization capabilityVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The system performs preliminary training on historical project data before actual operations begin. By pre-training machine learning models on past performance data, the system builds optimized planning capabilities in advance, enabling it to automatically generate improved workflows without requiring complex real-time decision-making during operations.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent implements feedback loops where actual operation results are fed back into the machine learning models to continuously improve predictions. This allows the system to learn from past performance and progressively optimize its planning accuracy, transforming static conventional methods into adaptive, self-improving systems.

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20250093544A1Machine learning and physics fusion modeling on run duration

Publication Date: 2025.03.20 SCHLUMBERGER TECH CORP

AI summary

Embodiments presented provide for modeling of wireline runs for hydrocarbon recovery operations. In embodiments, a run duration of wireline activities is split into a winch duration and a pass duration, wherein the pass duration is calculated using a machine learning model.