RBF Model Predicting TOC From NMR Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for estimating total organic carbon (TOC) in reservoirs are either expensive and time-consuming, such as laboratory tests, or impractical for large-scale applications, and nuclear logging methods often rely on assumptions that are not valid, especially when NMR logs include signals from organic matters in source rocks.
Innovation Solution
The use of a radial basis function (RBF) model applied to nuclear magnetic resonance (NMR) data to predict TOC values, incorporating inputs like NMR relaxation-time distributions, gamma ray logging data, and clay bound water values, allowing for real-time or near-real-time TOC estimation during drilling operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If laboratory mineralogy and fluid analysis using XRD and XRF are used to obtain TOC, then measurement precision is improved, but productivity deteriorates due to high cost and time consumption
Solution Approach 1:
The patent replaces mechanical/chemical laboratory analysis systems (XRD, XRF) with a nuclear magnetic resonance (NMR) logging system that uses electromagnetic fields to detect hydrogen distribution in rock pores. This substitution enables rapid TOC estimation during drilling operations without requiring physical sample extraction and laboratory testing, thereby resolving the contradiction between measurement precision and productivity
Solution Approach 2:
The patent creates a virtual model of TOC distribution by using NMR logging data to predict TOC values at different depths. Instead of physically testing numerous rock samples in the laboratory, the system generates a continuous TOC profile through electromagnetic measurement and computational prediction, achieving both speed and accuracy
2Productivity
If nuclear logging based mineralogy analysis is used to estimate TOC, then productivity is improved, but measurement precision deteriorates because not all elements can be resolved
Solution Approach 1:
The patent introduces NMR logging as an intermediary measurement method that indirectly estimates TOC through hydrogen detection in pore fluids. Instead of directly measuring carbon content (which requires resolving all elements), the system uses hydrogen distribution as a mediator to infer TOC values, achieving both speed and improved precision compared to traditional nuclear logging
Solution Approach 2:
The patent changes the measurement parameter from direct elemental composition analysis to hydrogen distribution detection. By measuring proton relaxation times and signal amplitudes in the NMR system, the method transforms the measurement approach to one that can be performed rapidly while providing accurate TOC estimation through the relationship between hydrogen content and organic carbon
3Productivity
If density and NMR logging measurements are combined to estimate TOC, then productivity is improved, but measurement precision deteriorates due to invalid assumptions about organic matter signals
Solution Approach 1:
The patent applies preliminary anti-action by using the NMR logging data to identify and correct for signals from organic matter before performing TOC estimation. Instead of assuming NMR logs contain no organic matter signals (which causes error), the system first detects these signals and then adjusts the calculation to eliminate their interfering effect, thereby improving measurement precision while maintaining productivity
Solution Approach 2:
The patent implements feedback by using the NMR logging measurements to provide information about organic matter presence, which then feeds back into the TOC estimation process. The system continuously refines the TOC calculation by incorporating real-time NMR data about hydrogen distribution and organic matter signals, improving accuracy without sacrificing the speed of field operations
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables accurate and efficient prediction of TOC values using NMR data, reducing the need for costly laboratory tests and improving the practicality of TOC estimation in field operations by providing a reliable log of TOC values as a function of measured depth.
Implementation Method 1
nuclear magnetic resonance (NMR) phenomena... The behavior of nuclei in presence of B0 and B1 has been correlated with formation rock and fluid properties such as the amount of hydrogen in pore space fluids
Data Source
AI summary
Systems, methods, and software for predicting total organic carbon (TOC) values are described. A representative method includes obtaining nuclear magnetic resonance (NMR) data and training a radial basis function (RBF) model based on the NMR data and measured total organic carbon (TOC) values. The method also includes obtaining subsequent NMR data and employing the trained RBF model to predict TOC values based at least in part on the subsequent NMR data. The method also includes storing or displaying the predicted TOC values.