Borehole Resonance Mode Machine Learning for Through-Tubing Cement Evaluation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional acoustic tools for cement bond evaluation in wellbores require the production tubing to be pulled out, making through-tubing cement evaluation (TTCE) challenging due to insufficient acoustic wave penetration, which complicates the assessment of cement bonding conditions behind the casing.
Innovation Solution
A machine learning-based approach using borehole resonance signals to generate 1-dimensional and 2-dimensional bonding indices and maps, employing acoustic logging tools with transmitters and receivers to analyze resonance modes and signals, and applying machine learning algorithms for feature extraction and prediction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional acoustic tools are used for cement bond evaluation, then cement bonding assessment can be performed, but the production tubing must be pulled out which complicates the evaluation process and reduces operational efficiency
Solution Approach 1:
The evaluation process is segmented into distinct functional components: acoustic signal generation, resonance mode excitation, signal processing, and machine learning analysis. This segmentation allows the system to perform cement bond evaluation through the tubing without requiring tubing removal, thereby improving operational efficiency while maintaining evaluation accuracy
Solution Approach 2:
Acoustic resonance modes act as an intermediary mechanism that enables non-contact evaluation of cement bonding. The resonance signals propagate through the tubing and cement sheath without requiring physical access or tubing removal, solving the operational complexity issue while providing accurate bonding assessment
2Measurement precision
If acoustic waves are transmitted through tubing for cement evaluation, then through-tubing evaluation is enabled, but acoustic wave penetration is insufficient which reduces measurement accuracy
Solution Approach 1:
The system utilizes mechanical vibration in the form of acoustic resonance modes to enhance wave penetration through the tubing and cement sheath. By exciting specific resonance modes, the acoustic energy is concentrated and amplified, enabling sufficient penetration to accurately evaluate cement bonding without requiring tubing removal
Solution Approach 2:
The system changes the acoustic parameters by selecting and adjusting specific resonance modes (frequency, mode shape) to optimize penetration and bonding assessment. This parameter adjustment allows the acoustic waves to effectively traverse the tubing and cement interface, improving measurement accuracy while maintaining through-tubing operation
3Measurement precision
If machine learning algorithms are applied to analyze resonance signals, then bonding index generation is improved, but computational complexity and processing time increase
Solution Approach 1:
The system performs preliminary signal processing actions including resonance mode identification, signal filtering, and feature extraction before the machine learning analysis. This preliminary action reduces the complexity of the input data to the ML algorithm, thereby reducing processing time while maintaining high bonding index accuracy
Solution Approach 2:
The system replaces complex mechanical signal analysis with machine learning algorithms that can efficiently process and interpret resonance signals. The ML models are trained to recognize bonding patterns, enabling accurate bonding index generation with reduced computational complexity compared to traditional signal processing methods
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 effective evaluation of cement bonding conditions within wellbores without the need to remove production tubing, providing accurate 1D and 2D bonding indices and maps, thereby improving the assessment of cement integrity.
Implementation Method 1
borehole resonance mode for cement evaluation using machine learning
Implementation Method 2
acoustic logging tools with transmitters and receivers to analyze resonance modes and signals
Data Source
AI summary
Systems and methods are provided for evaluation of the cement bonding condition in a wellbore based on borehole resonance mode using machine learning. An example method can include transforming the return signal into a resonance signal based on feature extraction of the return signal, determining a segment of the resonance signal in a time domain, and determining, via a machine learning model, a predicted borehole cement bonding based on the segment of the resonance signal. The example method can further include generating a bonding log based on the predicted borehole cement bonding.