Lithology Forecasting Ahead of the Bit Using Offset Well Logs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for predicting lithology and formation boundaries during drilling are inefficient and lead to inaccurate predictions due to the inability of wellsite geologists to manually integrate all relevant geological and drilling data, resulting in potential well control incidents.
Innovation Solution
A method and system utilizing machine learning models, specifically a recurrent neural network (RNN) and gradient boosting techniques, to forecast lithology and formation boundaries ahead of the drill bit by analyzing offset drilling and lithology logs, adjusting drilling parameters in real-time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning models are used to forecast lithology and formation boundaries, then prediction accuracy is improved, but device complexity increases
Solution Approach 1:
The system divides the prediction task into two separate machine learning models: a first model that forecasts drilling logs and a second model that predicts lithology and formation boundaries. This segmentation allows each model to specialize in specific aspects of the prediction, improving overall accuracy while managing complexity through modular architecture.
Solution Approach 2:
The forecasted drilling log serves as an intermediary between the raw drilling data and the final lithology prediction. The first machine learning model generates this intermediate forecast, which then becomes input for the second model that performs the actual lithology prediction, creating a staged processing approach.
2Reliability
If multiple machine learning models are trained and integrated, then prediction reliability is improved, but ease of operation deteriorates
Solution Approach 1:
The system merges multiple machine learning models into a unified forecasting system where the first model's output (forecasted drilling log) is integrated as input for the second model. This combination allows the system to leverage the strengths of different modeling approaches while presenting a single integrated solution to the user.
Solution Approach 2:
The system uses feedback from the forecasted drilling log to improve lithology predictions. The first model's predictions about drilling parameters provide feedback that helps the second model make more accurate lithology and formation boundary predictions, creating a refined iterative process.
Data Source
AI summary
A method includes drilling a wellbore in a current well. An interval of the wellbore comprises a first portion of the wellbore and a second portion of the wellbore. The method also includes obtaining an offset drilling log and an offset lithology log for a geologically similar interval in an offset well and training a first machine learning model, using the offset drilling log, to produce a first trained machine learning model. The method further includes producing, using the first trained machine learning model, a forecasted drilling log for the second portion of the wellbore in the current well, training a second machine learning model, using a gradient boosting machine learning technique, the forecasted drilling log, and the offset lithology log, to produce a second trained machine learning model, and producing, using the second trained machine learning model, a forecasted lithology log for the second portion of the current well.