Research on a borehole drilling efficiency prediction model based on bacterial foraging optimization stacking ensemble learning

Key Points

• This research aims to develop a predictive model for borehole drilling efficiency using advanced ensemble learning methods.
• Constructed a dataset from drilling efficiency data and factors like drilling depth and rock properties.
• Trained three heterogeneous base learners (XGBoost, LightGBM, MLP) in parallel with a bacterial foraging optimization algorithm.
• Employed support vector regression as a meta-learner to integrate predictions from base learners.
• Achieved R² values above 0.93 for individual learners after optimization.
• PCC values exceeded 0.97, indicating high correlation accuracy.
• Final model's PCC value approached 0.98, ensuring reliability for simulation needs.

Abstract