Dynamic Multi-Distribution Inversion Models for Wellbore Feature Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Inversion models used in wellbore operations are prone to misinterpretation due to varying sensitivity to data distribution curves, geology, and frequency, leading to inefficient or unsuccessful operations.

Innovation Solution

A dynamic multi-distribution model is generated by combining multiple resistivity inversion models with varying numbers of distribution curves, incorporating quality control factors, and selecting optimal sections for accurate geological feature representation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a single distribution curve is used to generate an inversion model, then the model generation is simple and fast, but the model accuracy and reliability are reduced due to misinterpretation of geological features

Engineering Contradiction:

Improveinversion model accuracyVSAvoidmodel complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The inversion model is segmented into multiple distribution curves, each representing different geological scenarios. Instead of using a single distribution curve, the system divides the inversion process into multiple segments (distribution curves) that collectively provide a more accurate and comprehensive representation of subsurface geology, resolving the contradiction between model accuracy and complexity by structured decomposition

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The inversion model uses a composite approach by combining multiple distribution curves into a unified model. This composite structure integrates information from different distribution scenarios, creating a more robust and accurate inversion model that overcomes the limitations of single-distribution approaches while managing complexity through systematic integration

Inventive Principle:

Principle #40Composite materials

2Reliability

If multiple distribution curves are used to generate an inversion model, then the model accuracy and geological feature representation are improved, but the computational complexity and processing time increase

Engineering Contradiction:

Improveinversion model reliabilityVSAvoidmodel generation time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by pre-defining multiple distribution curves and their parameters before the actual inversion process. This preparation work is done in advance, allowing the inversion model to quickly reference and apply these pre-configured distributions during execution, thereby improving reliability without proportionally increasing generation time

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The inversion model dynamically selects and applies appropriate distribution curves based on the specific geological context and data characteristics. This dynamic approach allows the system to adaptively use multiple distributions when needed while avoiding unnecessary computational overhead in scenarios where fewer distributions suffice, balancing reliability and processing time

Inventive Principle:

Principle #15Dynamics

3Difficulty of detecting and measuring

If inversion models are generated with high sensitivity to data variations, then the models can detect subtle geological features, but the models become prone to misinterpretation and errors in wellbore operations

Engineering Contradiction:

Improvegeological feature detection capabilityVSAvoidoperational reliability

Core Design Contradiction:

Difficulty of detecting and measuringVSReliability

Solution Approach 1:

Different distribution curves are assigned different local qualities or characteristics tailored to specific geological features or depth ranges. Each distribution curve is optimized for detecting particular types of features (e.g., hydrocarbon zones, water zones, rock types) with appropriate sensitivity levels, allowing the system to detect subtle features locally while maintaining overall operational reliability through specialized rather than uniformly high sensitivity

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

Multiple distribution curves act as intermediaries between the raw inversion data and the final geological interpretation. These distribution curves serve as mediating layers that process and filter the data, reducing the direct impact of data variations and noise while still enabling detection of genuine subtle geological features, thus improving operational reliability

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS20250283394A1Generating dynamic multi-distribution inversion models to facilitate wellbore operations

Publication Date: 2025.09.11 HALLIBURTON ENERGY SERVICES INC

AI summary

A system can provide a dynamic multi-distribution model to facilitate a wellbore operation. For example, the system can receive, from a downhole tool deployed in a wellbore during a drilling operation, resistivity data for a geological formation associated with an interval of the wellbore. The system can further execute a resistivity inversion algorithm to generate distribution outputs of a resistivity inversion model using the resistivity data. Additionally, the system can select a section of each of the distribution outputs. Each section of each distribution output can correspond with a different segment of the interval of the wellbore. The system can then generate a dynamic multi-distribution model. The dynamic multi-distribution model can include the sections selected from each of the distribution outputs. The system can further output, by a user interface, the dynamic distribution output, which can be used to adjust the drilling operation.