A Method for Optimizing Drilling Parameters Based on the Synergistic Optimization of Cost and Efficiency

Adjusting drilling parameters affects both the rate of penetration (ROP) and drill bit wear, making ROP improvement alone an inadequate measure of overall drilling economics. However, incorporating the trade-off between drilling speed and the associated operating and wear costs into parameter optimization remains challenging. This study proposes a framework for jointly optimizing drilling cost and efficiency by integrating a Transformer–bidirectional long short-term memory (BiLSTM) model with multi-objective Bayesian optimization. The framework addresses nonlinear dependencies in drilling sequences and the propagation of prediction errors into parameter selection. Self-attention captures dependencies among sampling positions within the input window, while BiLSTM extracts sequential patterns to estimate ROP. WOB and RPM serve as decision variables, and time-dependent operating costs, energy costs, and bit wear costs are incorporated into cost per unit drilled length. A bi-objective optimization model accounts for predictive uncertainty and engineering constraints, while a rolling strategy recommends parameter ranges derived from near-optimal Pareto solutions. Experimental results show that Transformer-BiLSTM outperforms the LSTM, BiLSTM, and Transformer baselines across all three-evaluation metrics. The mean absolute error (MAE) and root mean square error (RMSE) are 0.5812 and 0.9408 m/h, respectively, with a coefficient of determination (R2) of 0.9515. In the case study of Well A in Block M, the joint optimization scheme increases the average ROP from 53.28 to 64.14 m/h, representing a 20.39% improvement. The number of bit replacements decreases from three to two. Compared with the scheme that optimizes ROP alone, the joint optimization scheme reduces total costs by 36.73%. It also achieves a higher drilling rate than the scheme that minimizes cost alone. By balancing ROP and operating costs, the proposed framework supports drilling parameter selection that jointly addresses drilling schedule requirements and economic objectives.

Authors

Institutions

Publication Details

Journal
Processes
Published
2026-11-19
DOI
https://doi.org/10.3390/pr14223722
Primary Topic
Drilling and Well Engineering
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

A Method for Optimizing Drilling Parameters Based on the Synergistic Optimization of Cost and Efficiency

徐建明, Mingyi Hu, Liangjun Yan
Processes
Drilling and Well Engineering
article

A Method for Optimizing Drilling Parameters Based on the Synergistic Optimization of Cost and Efficiency

徐建明, Mingyi Hu, Liangjun Yan
article en

Abstract

Adjusting drilling parameters affects both the rate of penetration (ROP) and drill bit wear, making ROP improvement alone an inadequate measure of overall drilling economics. However, incorporating the trade-off between drilling speed and the associated operating and wear costs into parameter optimization remains challenging. This study proposes a framework for jointly optimizing drilling cost and efficiency by integrating a Transformer–bidirectional long short-term memory (BiLSTM) model with multi-objective Bayesian optimization. The framework addresses nonlinear dependencies in drilling sequences and the propagation of prediction errors into parameter selection. Self-attention captures dependencies among sampling positions within the input window, while BiLSTM extracts sequential patterns to estimate ROP. WOB and RPM serve as decision variables, and time-dependent operating costs, energy costs, and bit wear costs are incorporated into cost per unit drilled length. A bi-objective optimization model accounts for predictive uncertainty and engineering constraints, while a rolling strategy recommends parameter ranges derived from near-optimal Pareto solutions. Experimental results show that Transformer-BiLSTM outperforms the LSTM, BiLSTM, and Transformer baselines across all three-evaluation metrics. The mean absolute error (MAE) and root mean square error (RMSE) are 0.5812 and 0.9408 m/h, respectively, with a coefficient of determination (R2) of 0.9515. In the case study of Well A in Block M, the joint optimization scheme increases the average ROP from 53.28 to 64.14 m/h, representing a 20.39% improvement. The number of bit replacements decreases from three to two. Compared with the scheme that optimizes ROP alone, the joint optimization scheme reduces total costs by 36.73%. It also achieves a higher drilling rate than the scheme that minimizes cost alone. By balancing ROP and operating costs, the proposed framework supports drilling parameter selection that jointly addresses drilling schedule requirements and economic objectives.

ProcessesVol. 14(22)
Sinopec (China) (CN), Yangtze University (CN)
Drilling and Well Engineering
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon