Bayesian Optimization for Wellbore Completion Parameters

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Determining optimal completion parameters for wellbores is challenging due to unknown or changing environments, making it difficult to adjust parameters post-production initiation.

Innovation Solution

A system utilizing a neural network and Bayesian optimization to predict stimulation, production, and cost, incorporating physics-based models and sensor data to determine optimal completion parameters such as cluster spacing, perforation diameter, and proppant conductivity, which are used to plan and execute the completion stage of a wellbore.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If completion parameters are determined using traditional methods, then the process is simpler, but the accuracy and optimization of production parameters deteriorate due to unknown or changing wellbore environments

Engineering Contradiction:

Improveaccuracy of completion parametersVSAvoidcomplexity of optimization system

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system performs preliminary actions by collecting and analyzing wellbore environment data before completion operations begin. Bayesian optimization frameworks are established in advance to predict optimal completion parameters, allowing the system to proactively adapt to unknown or changing environments rather than reacting after problems occur.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements continuous feedback mechanisms where completion parameters are optimized based on real-time wellbore environment data. The Bayesian optimization framework updates predictions as new data becomes available, creating a closed-loop system that continuously refines completion parameter accuracy throughout the operation.

Inventive Principle:

Principle #23Feedback

2Productivity

If completion parameters are optimized using Bayesian optimization and machine learning, then production yield increases, but the computational time and processing complexity increase

Engineering Contradiction:

Improvehydrocarbon production yieldVSAvoidcomputational optimization time

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The system performs preliminary computations by pre-training machine learning models and establishing Bayesian optimization frameworks before completion operations begin. This allows the system to have optimization algorithms ready and waiting, reducing computational time during actual completion operations while still achieving high production yield optimization.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system implements dynamic optimization where the Bayesian framework adapts computations based on data availability and operational priorities. The system can adjust the level of optimization complexity in real-time, balancing computational effort against the need for precise completion parameters based on current operational context.

Inventive Principle:

Principle #15Dynamics

3Adaptability or versatility

If completion parameters are determined after production begins, then adjustments can be made to optimize production, but it becomes difficult or impossible to change parameters

Engineering Contradiction:

Improveability to adjust completion parametersVSAvoidtiming for parameter adjustment

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system performs preliminary determination of completion parameters using Bayesian optimization before production begins. By establishing the optimization framework in advance and predicting optimal parameters prior to completion operations, the system eliminates the need for post-production adjustments while still achieving adaptability through data-driven predictions.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system replaces physical reconfiguration mechanisms with computational optimization. Instead of requiring mechanical or operational changes to adjust completion parameters after production begins, the system uses machine learning and Bayesian optimization to predict and determine optimal parameters in advance, substituting computational intelligence for physical adaptability mechanisms.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS11493664B2Multi-objective completion parameters optimization for a wellbore using Bayesian optimization

Publication Date: 2022.11.08 LANDMARK GRAPHICS CORP

AI summary

A system for determining completion parameters for a wellbore includes a sensor and a computing device. The sensor can be positioned at a surface of a wellbore to detect data prior to finishing a completion stage for the wellbore. The computing device can receive the data, perform a history match for simulation and production using the sensor data and historical data, generate inferred data for completion parameters using the historical data identified during the history match, predict stimulated area and production by inputting the inferred data into a neural network model, determine completion parameters for the wellbore using Bayesian optimization on the stimulated area and production from the neural network model, profit maximization, and output the completion parameters for determining completion decisions for the wellbore.