ANN-Based Well Logging Tool for Formation EM Property Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current EM well-logging technologies face challenges in accurately characterizing formation resistivity and permittivity, especially in complex formations with directional drilling, due to computational intensity and poor convergence in inversion-based workflows, particularly for large standoffs and high formation resistivities.
Innovation Solution
The implementation of a system using artificial neural networks (ANNs) that predicts electromagnetic properties of drilling mud and formation by processing current measurements at multiple frequencies, eliminating the need for iterative inversion and enabling faster operation by employing a cascaded architecture of ANNs to directly predict mud and formation parameters, including tool standoff.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If inversion-based workflows are used to characterize formation resistivity and permittivity, then measurement precision is improved, but productivity deteriorates due to computational intensity and poor convergence
Solution Approach 1:
The patent pre-computes electromagnetic responses for a comprehensive library of formation models covering various resistivity and permittivity combinations before actual logging operations. This pre-computed library serves as a lookup table that enables rapid characterization during field operations without requiring real-time iterative inversion, thus resolving the contradiction between measurement precision and productivity
Solution Approach 2:
The patent creates simplified surrogate models that replicate the complex electromagnetic response characteristics of detailed formation models. These surrogate models are trained on pre-computed data and can quickly predict formation properties from measured responses, providing accurate characterization at fraction of the computational cost of full inversion workflows
2Measurement precision
If inversion-based workflows are used for large standoffs and high formation resistivities, then measurement precision is maintained, but reliability deteriorates due to poor convergence
Solution Approach 1:
The patent pre-computes electromagnetic responses specifically covering the challenging parameter space of large standoffs and high formation resistivities. By having pre-calculated responses for these difficult conditions, the system avoids iterative inversion that may fail to converge, ensuring reliable and accurate characterization even in challenging measurement scenarios
Solution Approach 2:
The patent introduces pre-computed electromagnetic response libraries and surrogate models as intermediaries between the measured data and formation properties. These intermediaries translate complex measurements into reliable formation characterizations without requiring direct iterative inversion, thereby improving convergence reliability while maintaining precision
3Measurement precision
If iterative inversion is used to predict mud and formation parameters, then measurement precision is improved, but loss of time increases due to computational intensity
Solution Approach 1:
The patent performs all computationally intensive iterative inversion operations in advance during an offline training phase, creating pre-computed lookup tables and trained surrogate models. During actual logging operations, parameter prediction becomes a simple lookup or evaluation task that delivers accurate results in real-time, eliminating the time loss associated with on-site iterative inversion
Solution Approach 2:
The patent replaces the mechanical iterative inversion process with computational models (surrogate models and lookup tables) that have been pre-trained. This substitution transforms a time-consuming iterative computational process into a rapid evaluation process, maintaining measurement precision while dramatically reducing the loss of time during 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
This approach significantly reduces computational time, achieving an order of magnitude faster prediction speed and improving accuracy by directly predicting mud and formation parameters without iterative inversion, thus enhancing the efficiency and reliability of EM well-logging in complex formations.
Implementation Method 1
The transmitter coil excites an alternating current (ac) inducing an alternating EM field that propagates/diffuses through the earth formation
Implementation Method 2
This EM field induces an electric current on the receiver coil. The electric current induced on the receiver is proportional to the conductivity of the formation
Data Source
AI summary
Methods and systems are provided that predict electromagnetic properties of drilling mud and a formation, which involve a logging tool that measures current injected into a measurement zone adjacent a sensor electrode at multiple frequencies. The measured currents at the multiple frequencies are processed to determine complex impedances for the sensor electrode at the multiple frequencies. The complex impedances are used to generate input data, which is supplied to a system of artificial neural networks (ANNs) that is configured to predict and output electromagnetic properties of the drilling mud and the formation within the measurement zone and possibly tool standoff based on the input data. The system of ANNs can employ a cascaded architecture of multiple ANNs. The electromagnetic properties or tool standoff predicted by the system of ANNs can be used to construct a borehole image over varying azimuth and depth.