Neural Network Hydraulic Fracture Geometry Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
There is uncertainty regarding hydraulic fracture geometry in subsurface formations due to vertical and lateral heterogeneity of rock properties, which affects the viability and successful production of subsurface resources.
Innovation Solution
A method is disclosed that involves obtaining data for a subsurface formation, generating hydraulic fracturing simulation data, transforming stress and leakoff coefficient data, and training convolution neural networks to predict hydraulic fracture height and geometry.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional hydraulic fracturing methods are used, then fracture treatment can be performed, but uncertainty regarding fracture geometry occurs due to rock property heterogeneity
Solution Approach 1:
The patent transforms input parameters (minimum horizontal stress data, leakoff coefficient data, plane strain Young's modulus data) through mathematical transformations before feeding them to neural networks. This parameter transformation approach enables the models to accurately capture fracture geometry despite rock property heterogeneity, resolving the contradiction between measurement precision and reliability.
Solution Approach 2:
The patent introduces trained convolutional neural network models as intermediary systems between raw formation data and fracture geometry determination. These neural network intermediaries process and interpret complex heterogeneous rock property data, providing reliable fracture geometry predictions that traditional direct methods cannot achieve.
2Measurement precision
If multiple simulations are run to account for heterogeneity, then more comprehensive data is obtained, but computational time and complexity increase
Solution Approach 1:
The patent performs preliminary action by training convolutional neural network models offline using extensive simulation data that accounts for rock property heterogeneity. Once trained, these models can rapidly predict fracture geometry without requiring repeated simulations, thus achieving high precision while minimizing real-time computational time loss.
Solution Approach 2:
The patent creates a simplified computational copy of the complex physical fracturing process through trained neural network models. This digital copy captures the essential physics and heterogeneity effects, allowing rapid prediction of fracture geometry without running computationally intensive physical simulations for each case.
Data Source
AI summary
A method is described for determining hydraulic fracture geometry. The method may be executed by a computer system. The method include training a first model that predicts a hydraulic fracture height for a hydraulic fracture using a first convolution neural network. The method includes training a second model that predicts a hydraulic fracture geometry for the hydraulic fracture using a second convolution neural network, the predicted hydraulic fracture height from the first trained model. Hydraulic fracture geometry may be determined for a target hydraulic fracture using the first and second trained models.