Quantitative estimate of rotary–percussion drilling efficiency in rocks

Oparin, V. N.; Timonin, Vadim; Karpov, V. N.

doi:10.1134/s1062739116061637

Cited by 14 publications

(1 citation statement)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nevertheless, data-driven models may overlook the underlying physical processes, potentially missing systemic causes of variations in efficiency. Oparin et al [7] determined the primary determinants for the drilling machine rate and presented a dimensionless energy criterion for rock destruction. However, their method primarily considers rock breakage mechanisms and does not fully account for the complexities associated with rotary drilling machines.…”

Section: Introductionmentioning

confidence: 99%

Prediction of Drilling Efficiency for Rotary Drilling Rig Based on an Improved Back Propagation Neural Network Algorithm

Jia,

Zhang,

Kong

et al. 2024

Machines

View full text Add to dashboard Cite

Accurately predicting the drilling efficiency of rotary drilling is the key to achieving intelligent construction. The current types of principle analysis (based on traditional interactive experimental methods) and efficiency prediction (based on simulation models) cannot meet the requirements needed for the efficient, real-time, and accurate drilling efficiency predictions of rotary drilling rigs. Therefore, we adopted a method based on machine learning to predict drilling efficiency. The extremely complex rock fragmentation process in drilling conditions also brings challenges to predicting drilling efficiency. Therefore, this article went through a combination of mechanism and data analysis to conduct correlation analysis and to clarify the drilling characteristic parameters that are highly correlated with drilling efficiency, and it then used them as inputs for machine learning models. We propose a rotary drilling rig drilling efficiency prediction model based on the GA-BP neural network to construct an accurate and efficient drilling efficiency prediction model. Compared with traditional BP neural networks, it utilizes the global optimization ability of a genetic algorithm to obtain the initial weights and thresholds of a BP neural network in order to avoid the defect of ordinary BP neural networks, i.e., that they easily fall into local optimal solutions during the training process. The average prediction accuracy of the GA-BP neural network is 93.6%, which is 3.1% higher than the traditional BP neural network.

show abstract

Section: Introductionmentioning

confidence: 99%