Well Integrity Inference From Legacy Data for CO2 Storage Risk
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for well integrity evaluation in depleted hydrocarbon reservoirs are inefficient and time-consuming, particularly for assessing CO2 storage sites, due to non-standardized legacy well data and the need for manual expert analysis, leading to prolonged lead times in CCS project delivery.
Innovation Solution
A backpropagation-enabled process is used to infer well integrity criteria from legacy well data through dependency-training and label-training, leveraging contextual relationships identified in a training well data set, enabling accurate and efficient risk assessment for CO2 storage sites.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual expert analysis is used to evaluate well integrity from legacy well data, then assessment accuracy can be maintained, but lead time for CO2 storage site risk assessment increases significantly (up to a year)
Solution Approach 1:
The patent replaces manual expert analysis (mechanical human cognitive process) with an automated machine learning system that uses backpropagation-enabled processes to evaluate well integrity criteria from legacy well data, thereby reducing lead time while maintaining assessment accuracy
Solution Approach 2:
The patent introduces a machine learning model as an intermediary between legacy well data and risk assessment outcomes, enabling automated extraction and evaluation of well integrity criteria without requiring direct manual expert intervention for each assessment
2Reliability
If extensive manual review of non-standardized legacy well data is performed, then comprehensive risk assessment can be achieved, but project delivery timelines are prolonged
Solution Approach 1:
The patent enables the system to perform self-service by automatically processing legacy well data through trained machine learning models that can independently evaluate well integrity criteria without requiring extensive manual review, thereby improving productivity while maintaining reliability
Solution Approach 2:
The patent applies preliminary action by pre-training machine learning models on legacy well data before deployment, enabling the system to rapidly assess new wells without requiring extensive manual preparation or review for each individual case
Data Source
AI summary
A method for inferring a well integrity criterion used for a CO2 storage site risk assessment of a subterranean formation uses a training well data set having a set of associated training labels. A backpropagation-enabled process is dependency-trained to identify contextual relationships between elements of the training well data set. The dependency-trained backpropagation-enabled process is label-trained using the training well data set and the associated training labels to assess a training well integrity criterion. The label-trained backpropagation-enabled process is used to compute a well integrity criterion in a non-training well data set.