Experimental Investigation and Optimization of the Penetration Rate of Large-Scale DTH Drilling in Copper Mines
Year: 2024
Type: article
Abstract: Optimizing the conditions of the drilling operation to achieve the maximum penetration rate (ROP) is essential as the initial step in mine-to-mill optimization. This study mainly focuses on optimizing the operational-scale drilling operation conditions of the drill wagon. For this purpose, two operational parameters (feed rate pressure and air flushing pressure) were identified as influential factors affecting the ROP. Various operational conditions were subsequently established using the Taguchi algorithm. Based on these conditions, drilling operations were conducted at the Anarak copper mine in Isfahan, Iran. Finally, the optimal operating conditions for different mine zones were developed. The results of the optimization showed that with the increase of copper grade and the decrease of rock strength, air flushing pressure has the most significant impact on the drilling ROP. Subsequently, a model was developed using three intelligent methods: Artificial Neural Network (ANN), Random Forest (RF), and Support Vector Machine (SVM). These models aimed to predict the ROP within the Python environment, utilizing a database comprising 170 drilling operations conducted at the Sarcheshmeh copper mine in Kerman, Iran. This database includes parameters that affect drilling ROP, including feed rate pressure, air flushing pressure, and geological strength index. To evaluate the performance of the developed models, three performance evaluation criteria (such as variance account for, normalized root mean square error, and coefficient of determination) were used based on the training database. The results showed that the RF model has a higher performance than the SVM and ANN model. Furthermore, the RF model was utilized for predicting the ROP in drilling operations at the Anarak copper mine. After conducting drilling operations in this mine and comparing the results, it was found that the RF model predicts the ROP with an accuracy of 83.8%. Therefore, the RF model can be confidently utilized in drilling projects with low uncertainty.
Source: Eurasian Mining
Authors: M. H. Kadkhodaei, R. Bagherpour, A. Nadri
Language: English
Topic: Tunneling and Rock Mechanics
Subfield: Civil and Structural Engineering
Field: Engineering
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4405332930