Training an Artificial Intelligence Model to Predict Bit Wear
By employing physics-based and data-driven filters to preprocess drilling data and project it into a latent space, the AI model effectively predicts drill bit wear, addressing the limitations of conventional methods and improving accuracy in hydrocarbon drilling operations.
US20260193972A1Pending Publication Date: 2026-07-09SAUDI ARABIAN OIL CO +1
View PDF 0 Cites 0 Cited by
Patent Information
- Application Number
- US19/011012
- Authority / Receiving Office
- US · United States
- Patent Type
- Applications(United States)
- Current Assignee / Owner
- Filing Date
- 2025-01-06
- Publication Date
- 2026-07-09
AI Technical Summary
Technical Problem
Conventional methods for predicting drill bit wear during hydrocarbon drilling are subjective, time-consuming, and limited by the availability of labeled data, failing to consider real-time formation and drilling conditions, and do not accurately account for similarities between offset and target wells.
Method used
A hybrid approach using physics-based and data-driven filters to preprocess drilling data from offset wells, projecting it into a latent space to uncover hidden patterns, and training an AI model with filtered data to predict bit wear accurately.
Benefits of technology
Enhances the accuracy of bit wear predictions by identifying relevant offset wells, allowing for timely adjustments in drilling operations and reducing the need for manual estimation.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure US20260193972A1-D00000_ABST
Abstract
A computer implemented method that enables training an artificial intelligence model to predict bit wear is described. The method includes applying multiple filters to a data pool associated with offset wells, wherein the filters comprise physics-based filters and data driven filters. A training set of relevant offset wells is created from a first set of relevant offset wells and a second set of relevant offset wells. An artificial intelligence model is trained to predict bit wear using the training set of relevant offset wells.
Need to check novelty before this filing date? Find Prior Art