Drill Bit Selection Architecture Using Regression Models

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

The existing methods for selecting drill bits are inefficient and require significant expertise and time, often leading to a fragmented product line that is difficult to manage, as they rely on local experts to simulate a handful of bits in a given environment and do not effectively account for the specific geology and drilling methods.

Innovation Solution

An architecture that uses regression models and machine learning to calculate a performance index for drill bits based on bit design, formation information, and historical data, allowing for the selection of the best-fit drill bits for specific applications and geographies, reducing the need for extensive expertise and time.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If local experts manually simulate drill bits in a given environment, then selection accuracy depends on expert knowledge, but the process requires significant time and expertise while leading to fragmented product line management

Engineering Contradiction:
Improvedrill bit selection accuracyVSAvoidselection time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces manual expert simulation with an automated computer-based system that uses regression models and machine learning algorithms to calculate performance indices. This substitution eliminates the need for expert mechanical simulation while providing consistent, reproducible results across different drill bit evaluations.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system transforms the selection process by changing from qualitative expert judgment to quantitative parameter-based evaluation. Multiple performance parameters (RPM, torque, penetration rate, vibration) are measured and processed through regression models to generate comprehensive performance indices, enabling objective comparison of drill bits.

Inventive Principle:
Principle #35Parameter changes

2Adaptability or versatility

If multiple drill bit designs are evaluated manually, then comprehensive comparison is possible, but the complexity of managing fragmented product line increases

Engineering Contradiction:
Improvedrill bit design varietyVSAvoidproduct line management complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent creates a universal evaluation system that handles multiple drill bit types (fixed cutter, roller cone, impregnated, coring) through a single integrated platform. The system uses standardized performance parameters and regression models that work across all drill bit varieties, eliminating the need for separate evaluation methods for each type.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system manages product line complexity by transforming diverse drill bit characteristics into standardized quantitative parameters. All drill bit designs are evaluated using the same performance indices and regression models, enabling systematic comparison and management regardless of drill bit type or complexity.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If traditional drill bit selection methods are used, then expertise-dependent decisions are made, but the ability to account for specific geology and drilling methods is limited

Engineering Contradiction:
Improveselection reliabilityVSAvoidgeology-specific optimization
Core Design Contradiction:
ReliabilityVSAdaptability or versatility

Solution Approach 1:

The patent performs preliminary characterization of formation properties and drilling conditions before drill bit selection. The system pre-processes geology data, wellbore information, and drilling parameters to create optimized input sets for the regression models, enabling tailored recommendations for specific geological conditions before the actual selection process.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system incorporates feedback mechanisms where historical drilling data and performance measurements are continuously fed back into the regression models. This allows the system to learn from past performance and improve its recommendations for specific geologies and drilling methods over time, increasing both reliability and adaptability.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3931423B1System and architecture for comparing and selecting a drill bit design
Publication Date: 2025.01.08 SERVICES PETROLIERS SCHLUMBERGER SA
  • EP3931423B1 patent drawingFigure 1
  • EP3931423B1 patent drawingFigure 2
  • EP3931423B1 patent drawingFigure 3

AI summary

Systems and methods discussed herein relate to applying models to downhole tool records to identify, sort, and display a subset of available downhole tool records with index information as defined by the models that have a desired relationship with a received input. The received input may be indicative of a particular device, device family, or set of features for a downhole tool. The methods may include identifying, from a set of design information stored in computer-storage media, one or more downhole tool records that correspond to the received input, applying one or more index models to the identified one or more downhole tool records, applying one or more local models to the identified one or more downhole tool records, and displaying at least some of the one or more downhole tool records, with index information as defined by the one or more index and local models.