基于自组织神经网络与元启发式算法的地表爆破振动预测方法.

A method for predicting surface vibrations induced by blasting in open-pit mines is proposed, based on Self-organizing Neural Network (SONN) and several commonly used metaheuristic algorithms to improve the prediction accuracy of the SONN model. These algorithms include Manta Ray Foraging Optimization (MRFO), Hunger Games Search (HGS), Aquila Optimization (AO), and Naked Mole-rat Algorithm (NMRA). The k-fold cross-validation was employed to determine the optimal parameters for the algorithms, which were then used to retrain the model for predicting blast-induced surface vibrations. A case study of a domestic open-pit mine was conducted to validate the effectiveness of the proposed method. The research results indicate that all 4 models accurately predict blast-induced surface vibrations. Among them, the prediction accuracy and reliability are ranked from highest to lowest as follows: MRFO-SONN model>HGS-SONN model>NMRA-SONN model>AO-SONN model. The MRFO-SONN model is recommended for predicting surface vibrations induced by blasting activities. [ABSTRACT FROM AUTHOR]