Probabilistic Fracturing Model for Breakdown Pressure Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional hydraulic fracturing simulators fail to accurately predict subsurface formation breakdown pressure due to model simplifications, leading to inadequate fracturing liquid injection and potential need for more powerful pumps, resulting in increased costs and delays.
Innovation Solution
A probabilistic model that accounts for uncertainties in well structure, subsurface formation properties, and modeling parameters using a stochastic finite element approach and Monte Carlo simulations to determine upper and lower bounds of breakdown pressure, integrated with a deterministic computational model to optimize pump schedule and equipment sizing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If conventional hydraulic fracturing simulators use simplified models to predict breakdown pressure, then the device complexity is reduced and ease of operation is improved, but the measurement precision and reliability of breakdown pressure prediction deteriorate
Solution Approach 1:
The patent transforms the deterministic model into a probabilistic model by changing the parameters from fixed values to statistical distributions. This allows the model to account for uncertainties in rock properties, wellbore geometry, and operational parameters, thereby improving prediction accuracy without requiring a complete redesign of the computational framework.
Solution Approach 2:
The patent introduces Monte Carlo simulation as an intermediary method between the simplified deterministic model and the complex real-world system. This intermediary approach uses random sampling to explore the range of possible outcomes, providing a bridge that maintains computational efficiency while improving prediction reliability through probabilistic analysis.
2Productivity
If conventional simulators use simplified models for breakdown pressure prediction, then the computation time is reduced and productivity is improved, but the reliability of fracturing treatment design deteriorates
Solution Approach 1:
The patent performs preliminary probabilistic analysis by defining statistical distributions for all input parameters before running the main simulation. This preliminary preparation allows the Monte Carlo method to efficiently sample from these distributions, maintaining computational speed while ensuring that uncertainty is systematically accounted for in the breakdown pressure prediction.
Solution Approach 2:
The patent implements feedback through the iterative Monte Carlo simulation process, where each iteration provides information about the range and probability of possible breakdown pressures. This feedback mechanism allows the model to converge on a reliable prediction range while maintaining computational efficiency through the use of statistical sampling rather than exhaustive analysis.
3Device complexity
If deterministic models are used to calculate breakdown pressure, then the device complexity and computation time are reduced, but the ability to account for uncertainties in rock properties, wellbore structure, and modeling parameters deteriorates
Solution Approach 1:
The patent makes the computational model universal by integrating both deterministic calculation and probabilistic analysis capabilities into a single framework. The model can handle both fixed-parameter deterministic scenarios and uncertainty-rich probabilistic scenarios, making it adaptable to various levels of data availability and uncertainty without requiring separate modeling approaches.
Solution Approach 2:
The patent introduces dynamics to the model by allowing parameters to vary according to their statistical distributions rather than remaining fixed. This dynamic approach enables the model to adapt to different uncertainty scenarios, where parameters can take on different values in different Monte Carlo iterations, thereby capturing the inherent variability and uncertainty in the fracturing system.
Data Source
AI summary
Techniques for hydraulic fracturing a subsurface formation can include using a deterministic model to simulate a deviated well comprising a casing and at least one perforation tunnel. The techniques can include determining a different statistical distribution for each of one or more parameters to the deterministic model. The statistical distributions can be determined from the log data along the measured depth within the wellbore. The techniques can include probabilistically solving the deterministic model to determine a mean and a standard deviation of breakdown pressures along the measured depth within the wellbore. The techniques can include processing the mean and the standard deviation of breakdown pressures to determine an upper bound breakdown pressure based on a level of confidence. The techniques can include drilling and completing a deviated well based on the determined upper bound breakdown pressure and injecting hydraulic fluid to cause hydraulic fracturing of the subsurface formation.