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

Borehole drilling efficiency is a key parameter in construction progress simulation of rock-fill dam quarry excavation， and its prediction accuracy directly affects the reliability of the simulation model. To address the issue of low prediction accuracy efficiency in existing mathematical methods， the difficulty of single learners meeting simulation accuracy requirements， and insufficient local search refinement in hyperparameter tuning within ensemble learning research， this paper proposes a borehole drilling efficiency prediction model based on Bacterial Foraging Optimization for Stacking ensemble learning. First， a dataset was constructed using on-site drilling efficiency data from a rock-fill dam as the target variable and its influencing factors （e.g.， borehole depth， rock properties， altitude） as feature variables. Second， three heterogeneous base learners （XGBoost， LightGBM， and MLP） were trained in parallel， and the Bacterial Foraging Optimization algorithm—simulating chemotaxis and reproduction—was introduced to iteratively optimize each base learner’s hyperparameters by tracking the R² curve in real time， ensuring stable “meta-features” output. Finally， the base learners’ predictions were input to a Support Vector Regression （SVR） meta-learner； by integrating complementary information from multiple models， the ensemble prediction was obtained while suppressing bias and variance. Experimental results show that after Bacterial Foraging Optimization， each base learner’s R² can exceed 0.93 and PCC are all above 0.97； the ensemble model’s learning curve over the full sample set is smooth and stable， residual analysis indicates residuals are evenly distributed around the zero-mean line， and the final PCC approaches 0.98， meeting the requirements of construction process simulation.