NMR Logging Data Processing via Dual-Step Inversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current 3D NMR inversion methods for deriving T1, T2, and D distributions from NMR data are computationally expensive and time-consuming, making them impractical for field data interpretation in wellbore environments due to the large number of unknowns and under-determined problems, especially when requiring 3D inversion at each depth level.
Innovation Solution
The method employs a dual step independent inversion approach, applying physical constraints to reduce the matrix size by using a 2.5D model for D-T2 and D-T1 maps, allowing for separate and independent derivation of D-T1 and D-T2 maps from NMR data without full 3D inversion, significantly reducing computational complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full 3D inversion is performed to derive T1, T2, and D distributions, then measurement precision is improved, but computational complexity and processing time increase significantly
Solution Approach 1:
The patent segments the 3D inversion problem into two separate 2.5D inversion problems. First, a D-T2 map is derived from NMR data, then a D-T1 map is derived independently. This segmentation reduces the computational complexity by avoiding the need to simultaneously invert all three parameters (T1, T2, D) in a single 3D problem, while still providing accurate distributions through sequential independent inversions.
Solution Approach 2:
The patent transitions from a 3D inversion approach to a 2.5D approach by fixing one parameter dimension at a time. Instead of simultaneously inverting T1, T2, and D in three dimensions, the method first inverts for D and T2 (2.5D), then independently inverts for D and T1 (2.5D). This dimensional reduction significantly decreases computational complexity while maintaining measurement precision.
2Measurement precision
If 3D inversion is performed at each depth level, then measurement precision is improved, but processing time increases making it impractical for field data interpretation
Solution Approach 1:
The patent applies segmentation by dividing the inversion process into two separate 2.5D inversion steps instead of one 3D inversion step at each depth level. This segmentation reduces the computational burden per depth level, making the overall processing time acceptable for field data interpretation while maintaining accuracy through the sequential derivation of D-T2 and D-T1 maps.
Solution Approach 2:
The patent reduces the inversion from 3D to 2.5D by independently deriving maps with fewer parameters at each step. This dimensional change significantly reduces processing time at each depth level, making the method practical for field applications where rapid interpretation is essential, while still providing accurate T1 and T2 distributions through the independent inversions.
3Measurement precision
If the number of unknowns is increased to fully characterize fluid properties, then measurement precision is improved, but the problem becomes under-determined and computationally intractable
Solution Approach 1:
The patent segments the set of unknowns into two separate inversion problems. Instead of simultaneously solving for three unknowns (T1, T2, D) which creates an under-determined problem, the method first solves for two unknowns (D, T2) to create a D-T2 map, then independently solves for another two unknowns (D, T1) to create a D-T1 map. This segmentation makes each inversion problem well-determined and computationally tractable while still achieving comprehensive fluid characterization.
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
This approach provides accurate and efficient derivation of T1 and D-T1 information, improving fluid-typing capabilities and identifying fluids like heavy oil that cannot be distinguished by D-T2 or D-T1 alone, with improved fidelity of the T1 spectrum and reduced processing time.
Implementation Method 1
nuclear magnetic resonance (NMR) data acquired for a formation from within a subterranean wellbore
Data Source
AI summary
Apparatus, method and system for processing and interpreting nuclear magnetic resonance (NMR) data acquired for a formation from within a subterranean wellbore that includes independently obtaining D−T1 and D−T2 from the same set of NMR data using a dual step independent 2D inversion method that provides adequate resolution in all dimensions for T1, T2, and D distributions.