Mud-Weight Window Calculation Using Hybrid Physics-AI Search Refinement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing physics-based models for calculating the mud-weight window in drilling operations are time-consuming, while AI solutions lack accuracy, and there is a need for a method that combines the speed of AI with the accuracy of physics-based methods to determine a reliable mud-weight window.
Innovation Solution
A hybrid physics-AI method that uses a computer processor to determine fracture and collapse mud weights through physics-based and AI methods, trains an AI network, and refines search windows to predict accurate mud-weight boundaries using a neural network.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If physics-based models are used to calculate the mud-weight window, then accuracy is improved, but calculation time increases significantly
Solution Approach 1:
The patent applies preliminary action by pre-training an AI network on a comprehensive dataset generated from physics-based models. This training phase performs the computationally intensive work in advance, allowing the trained AI model to rapidly predict mud-weight windows during actual drilling operations without requiring real-time physics-based calculations, thus achieving both speed and accuracy
Solution Approach 2:
The patent uses an AI network as an intermediary between the complex physics-based models and the real-time drilling operations. The AI model learns the underlying patterns from physics-based simulations and serves as a computationally efficient mediator that provides accurate predictions without requiring direct execution of complex physics calculations during operations
2Productivity
If AI solutions are used to calculate the mud-weight window, then calculation speed is improved, but accuracy cannot be guaranteed for all input data
Solution Approach 1:
The patent performs preliminary action by generating a comprehensive training dataset using physics-based models before deploying the AI system. This pre-computed dataset captures the full range of possible input conditions and their corresponding accurate physics-based results, enabling the AI model to learn accurate mappings across all scenarios rather than just average cases
Solution Approach 2:
The patent implements feedback by using physics-based model results as ground truth labels during the AI training process. The AI model learns from the feedback provided by accurate physics-based calculations, adjusting its parameters to minimize the difference between its predictions and the physics-based reference values, thereby ensuring accuracy is embedded in the AI model
Data Source
AI summary
A system and method for determining a mud-weight window is disclosed. The method includes determining, for each of a plurality of borehole-formation training models, a fracture mud weight using a first physics-based method and a collapse mud weight using a second physics-based method, and training, using the plurality of borehole-formation training models, an artificial intelligence (AI) network to predict the fracture and collapse mud weights for a borehole formation model. The method further includes determining, using the trained AI network, an AI fracture and an AI collapse mud weight from an observed borehole-formation model, and a search window surrounding each AI mud weight. The method further includes predicting a final fracture mud weight within the fracture search window using the first physics-based method and a final collapse mud weight within the collapse search window using the second physics-based method, and determining the mud-weight window bounded by the two final mud weights.