Prediction of the mean fragment size in mine blasting operations by deep learning and grey wolf optimization algorithm

Kai Rong,Xuan Xu,Haibo Wang,Jun Yang

DOI: https://doi.org/10.1007/s12145-024-01313-7

2024-05-08

Earth Science Informatics

Abstract:Mean fragment size is an important index in mine blasting since it significantly influences downstream work efficiency. Therefore, it is necessary to carry out the characteristic particle size prediction of rock fragments. Aiming at this topic, this study presents convolutional neural network (CNN), multilayer perceptron (MLP) and grey wolf optimization (GWO) algorithm, for predicting mean fragment size. The blasting data from multiple open pit mines is collected and augmented, and a large-scale database is established to conduct various parametric studies for both CNN and MLP models to obtain the best one. Then, the GWO algorithm is used to optimize the training process of the above model with the best training and validation performance to further improve the model performance. The results show that the best CNN model has a higher capacity in predicting mean fragment size than the MLP model and the GWO optimization algorithm can further improve the model performance. The model performance can be evaluated by four statistical indices, including mean square error (MSE), root mean square error (RMSE), prediction accuracy and coefficient of determination (R 2 ). The GWO-CNN model and data augmentation method proposed in this study can be introduced as an applicable method for estimating the mean fragment size in mine blasting.

geosciences, multidisciplinary,computer science, interdisciplinary applications