Hydraulic Fracturing Parameter Prediction with Injection Data and ML

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional hydraulic fracturing processes in low-permeability reservoirs are time-consuming and inefficient due to the need for multiple diagnostic pumping and analysis steps, such as minifrac procedures, which can lead to suboptimal pumping and equipment usage, and prolonged operational times.

Innovation Solution

A machine learning predictive model is employed to analyze injection test data, allowing for the direct prediction of hydraulic fracturing parameters without the need for minifrac procedures, optimizing the fracturing operation by utilizing a pre-trained model to determine optimal control parameters.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional diagnostic pumping and analysis steps (minifrac procedures) are performed to determine hydraulic fracturing parameters, then measurement precision is improved, but loss of time increases

Engineering Contradiction:

Improveprediction accuracy of fracture design parametersVSAvoidoperational time for fracturing preparation

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary injection tests and data collection before the main fracturing operation, using this advance information to train and apply machine learning models that predict fracturing parameters. This preliminary action eliminates the need for time-consuming minifrac procedures immediately before the main operation, reducing operational time while maintaining prediction accuracy through pre-gathered data analysis

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces conventional mechanical diagnostic procedures (minifrac pumping tests) with a machine learning-based predictive system. Instead of performing physical pumping and pressure analysis steps, the system uses trained ML models that process injection test data to directly predict fracture design parameters, substituting computational analysis for mechanical diagnostic operations

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Manufacturing precision

If multiple diagnostic pumping steps are performed to optimize fracturing parameters, then manufacturing precision is improved, but device complexity increases

Engineering Contradiction:

Improveprecision of fracturing parameter optimizationVSAvoidcomplexity of diagnostic equipment and procedures

Core Design Contradiction:

Manufacturing precisionVSDevice complexity

Solution Approach 1:

The machine learning system serves multiple functions: it predicts fracture geometry, optimizes pumping parameters, and analyzes injection test data all through a single integrated platform. This universal system replaces multiple specialized diagnostic tools and procedures, reducing equipment complexity while maintaining or improving parameter optimization precision through the multi-functional ML model

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Reliability

If conventional minifrac procedures are used to determine operational parameters, then reliability of fracturing design is improved, but productivity decreases

Engineering Contradiction:

Improvereliability of fracture design parametersVSAvoidoperational efficiency of fracturing process

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The system replaces mechanical minifrac procedures with machine learning-based prediction, maintaining reliability by using trained models that analyze injection test data to accurately predict fracture design parameters. This substitution eliminates the need for additional pumping and pressure testing steps, significantly improving productivity by reducing operational time while maintaining design reliability through data-driven predictions

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The machine learning system uses historical injection test data and operational results to continuously improve and self-calibrate its predictions. The system serves itself by learning from past performance and automatically optimizing future predictions, maintaining reliability without requiring additional manual diagnostic procedures or expert intervention, thereby improving overall productivity

Inventive Principle:

Principle #25Self-service

Data Source

PatentUS20250237131A1Systems and methods for predicting hydraulic fracturing design parmaters based on injection test data and machine learning

Publication Date: 2025.07.24 SCHLUMBERGER TECH CORP

AI summary

Systems and methods presented herein include systems and methods for receiving data relating to an injection/falloff test performed in a well in fluid communication with a subterranean reservoir; determining operational parameters of a hydraulic fracturing operation using at least a portion of the data; applying the operational parameters to a pre-trained machine learning predictive model to determine an optimal set of control parameters; and issuing one or more commands relating to the control parameters to optimize the hydraulic fracturing operation on the subterranean reservoir.