Casing Wear Modeling Using Discrete Inversion for Drilling Design
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for predicting casing wear, riser wear, and friction factors in drilling operations often result in overdesign of casing strings, leading to unnecessary capital investment and potential well abandonment, due to inherent uncertainties in casing wear estimation.
Innovation Solution
Data-driven models using discrete inversion techniques are developed to update casing wear, riser wear, and friction factor models, allowing for more accurate predictions by training models with available data from current or historical wells, employing linear and non-linear inversion methods to estimate wear factors and friction factors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional casing wear prediction methods are used, then casing integrity is maintained through overdesign, but capital investment increases unnecessarily
Solution Approach 1:
The patent transforms the casing wear prediction approach by changing the fundamental parameters used in prediction models. Instead of relying on traditional empirical methods with high uncertainty, the invention introduces discrete inversion techniques that calculate wear based on actual measured data from drilling operations. This parameter transformation enables accurate prediction of wear factors and friction coefficients, allowing engineers to determine minimum required casing specifications rather than using excessive safety margins, thus reducing unnecessary material usage while maintaining integrity
Solution Approach 2:
The patent replaces traditional mechanical/empirical prediction methods with computational mathematical models. By substituting physical trial-and-error approaches with discrete inversion algorithms that process drilling data, the system achieves precise wear predictions. This substitution eliminates the need for conservative overdesign that characterized traditional mechanical engineering approaches, enabling optimal casing design with reduced material quantity
2Device complexity
If traditional casing wear prediction methods are used, then engineering simplicity is maintained, but prediction accuracy deteriorates
Solution Approach 1:
The patent introduces discrete inversion techniques as an intermediary computational layer between raw drilling data and casing wear predictions. This intermediary process transforms measured drilling parameters into accurate wear factor and friction coefficient estimates, which then feed into refined prediction models. The discrete inversion acts as a mathematical mediator that bridges simple data collection and complex wear prediction, achieving high measurement precision while maintaining reasonable model complexity through systematic computational steps
3Measurement precision
If discrete inversion techniques are applied, then prediction accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent segments the complex casing wear prediction problem into distinct computational stages: first applying discrete inversion to calculate wear factors from drilling data, then separately computing friction coefficients, and finally integrating these into the wear prediction model. This segmentation of the computational process breaks down the complex mathematical operations into manageable steps, improving wear factor estimation accuracy while keeping computational complexity organized and controllable through systematic decomposition
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Predicting casing wear, riser wear, and friction factors in drilling operations may be achieved with data-driven models that use discrete inversion techniques to updated casing wear models, riser wear models, and/or friction factor models. For example, a method may applying a linear inversion technique or a nonlinear inversion technique to one or more parameters of at least one of a casing wear model, a riser wear model, or a friction factor model using historical data from a previously drilled well as input data to produce at least one of an updated casing wear model, an updated riser wear model, or an updated friction factor model, respectively; and implementing the at least one of the updated casing wear model, the updated riser wear model, or the updated friction factor model when designing and/or performing a drilling operation.