Abstract
The utilization factor of blasthole is a crucial indicator of the effectiveness of blasting in rock roadways. A significant value means that the explosive energy is fully utilized, the single-cycle advance is high, and the excavation rate is fast. A good blasting programme is a prerequisite for improving the utilization rate and predicting the utilization rate before blasting operations can verify the feasibility of the blasting programme. Firstly, a database of rock roadway blasting covering different geological and production conditions is established. Secondly, error analysis and the Gini coefficient method are used to weight the characteristic variables, quantify the importance of the variables and identify eight key indicators affecting the blasting hole utilization rate. Then, a random forest algorithm-based model for predicting utilization factor of blasthole is proposed, and the results of the model on the test set are: root mean square error (RMSE) is 0.0137, mean absolute error (MAE) is 0.0087, and coefficient of determination (R2) is 0.905. The performance of this method is compared with that of the neural network and support vector machine models on the test sets to verify the superiority of the random forest algorithm. Finally, to verify the generalization ability and practicality of the random forest prediction model, the model is applied to the rock roadway blasting construction of Gu Bei coal mine in Anhui Province. The results show that R2 is 0.913, so the model is reliable and accurate, which can meet the actual engineering requirements and lay the foundation for the promotion and application of this technology.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have