Drilling Dysfunction Prediction via Machine Learning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Drilling dysfunctions in unconventional subsurface wellbores, such as stuck pipes and wellbore instability, are often detected reactively, leading to decreased efficiency and increased operational costs due to the exhaustive and error-prone process of monitoring real-time drilling data by engineers.

Innovation Solution

A method and system that convert time-series real-time drilling data into dysfunction predictions using a machine-learning model layered on top of a deep learning model, processing data to determine rig states and trends, and generating indicators for predicted dysfunction severity, allowing for automated monitoring and visualization on user devices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If engineers visually monitor streaming output of real-time drilling data across multiple displays, then drilling dysfunction can be detected, but the process is exhaustive, error-prone, and decreases drilling efficiency

Engineering Contradiction:

Improvedetection accuracyVSAvoiddrilling efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent replaces the mechanical/visual monitoring system with an automated computer-based monitoring system. The system uses processors to automatically analyze drilling data streams, detect dysfunctions, and generate alerts, eliminating the need for engineers to manually monitor multiple displays while improving both reliability and productivity

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The monitoring system performs self-analysis of drilling data without requiring continuous human intervention. The automated system independently processes data streams, identifies dysfunction patterns, and generates notifications, allowing the system to serve itself in the detection process while freeing engineers for other tasks

Inventive Principle:

Principle #25Self-service

2Reliability

If engineers visually monitor real-time drilling data, then dysfunction detection is possible, but human error increases due to missing fluctuations and misinterpretation

Engineering Contradiction:

Improvedetection accuracyVSAvoiddata interpretation accuracy

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The patent replaces human visual monitoring with automated computational analysis. The system uses processors to objectively analyze drilling data streams, eliminating human errors in detecting data fluctuations and interpreting monitoring information while maintaining high detection accuracy

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Reliability

If reactive action is taken to remedy drilling dysfunctions, then the dysfunction is addressed, but drilling efficiency sharply decreases and operational expenses increase

Engineering Contradiction:

Improvedysfunction remediationVSAvoiddrilling efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent enables preliminary detection and alerting of drilling dysfunctions before they fully manifest or cause severe problems. The automated system continuously monitors data streams and generates early warnings, allowing operators to take preventive actions that maintain drilling efficiency while ensuring reliable dysfunction remediation

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system provides continuous feedback through automated alerts and notifications when dysfunctions are detected. This real-time feedback mechanism enables operators to respond promptly to drilling issues, maintaining both reliable dysfunction addressing and high drilling efficiency by preventing minor issues from escalating

Inventive Principle:

Principle #23Feedback

Data Source

PatentUS20230111036A1Integrated drilling dysfunction prediction

Publication Date: 2023.04.13 EXXONMOBIL TECHNOLOGY & ENGINEERING CO

AI summary

A computer system, computer, and method for converting time series real-time drilling data to a drilling dysfunction prediction, utilizing machine learning layered on top of deep learning with data processing and trend analysis therebetween.