Resistivity Logging Invasion Correction via Random Forest Inversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for invasion correction in resistivity logging, such as chart correction and electrical logging inversion, face challenges including inaccurate calculation of invasion radius, limited consideration of dynamic invasion processes, and high operational complexity, especially in tight sandstone gas reservoirs with low porosity and low permeability.
Innovation Solution
The proposed method combines forward modeling and inversion techniques using a random forest method to calculate a correction coefficient, merging multiple charts into a single set to determine formation conditions and output corrected resistivity values, while controlling parameters under complex borehole conditions to ensure accurate invasion correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If chart correction method is used to calculate Rt value according to RLLD/RLLS and Rxo, then correction effect on low salinity mud filtrate invasion is good, but correction effect on high salinity mud filtrate invasion is not ideal
Solution Approach 1:
The patent changes the parameters used in correction by introducing a new parameter K (correction coefficient) that is calculated based on multiple factors including salinity ratios. Instead of using fixed chart-based corrections, the method dynamically adjusts parameters according to actual invasion conditions, allowing accurate correction across both low and high salinity scenarios
Solution Approach 2:
The patent combines multiple correction approaches into a composite correction formula that integrates chart correction results with inversion-based correction coefficients. This composite method leverages the strengths of both approaches to achieve accurate correction across diverse salinity conditions
2Measurement precision
If inversion method is used to obtain Rt value and invasion radius, then accurate determination can be achieved, but operational complexity increases and huge amount of sample data is required
Solution Approach 1:
The patent applies partial inversion by using inversion methodology only to calculate the correction coefficient K, rather than performing complete inversion to obtain all formation parameters. This partial application reduces computational complexity and data requirements while maintaining accuracy in the specific task of invasion correction
Solution Approach 2:
The patent performs preliminary classification of invasion types before applying correction, using simple criteria based on logging curve characteristics. This preliminary action simplifies the subsequent correction process by allowing different correction strategies for different invasion scenarios
3Productivity
If existing invasion radius calculation methods (Tan's estimation, volumetric method, skin factor method) are used, then calculation can be performed, but dynamic change features of invasion process over time are not considered
Solution Approach 1:
The patent introduces dynamic elements by calculating invasion radius as a function that evolves with invasion time, rather than using static estimation methods. The correction coefficient K incorporates temporal dynamics of the invasion process, allowing the model to reflect how invasion progresses over time and affects resistivity measurements at different stages
Data Source
AI summary
Disclosed is an invasion correction method and system for resistivity logging. The method includes the steps of: analyzing mud filtrate invasion features based on an actual resistivity logging curve; determining an invasion mechanism, and performing forward modeling on a resistivity logging response; performing inversion by a random forest method to form a correction chart; and outputting a formation resistivity to finally form the invasion correction method and system for resistivity logging. In the present disclosure, double lateral correction under an invasion condition is accomplished, while a chart method and an inversion method are combined, ensuring the accuracy and operating rate of correction.