Abstract Blasting remains as an economical and reliable excavation technique, but there are some environmental shortcomings such as the control of blast-induced vibration. The impacts of vibration over surrounding communities in a blast area have been investigated for decades and researchers have been using a myriad of empirical predictive attenuation equations. These models, however, may not have satisfactory accuracy, since parameters associated to geomechanical properties and geology affect the propagation of seismic waves, making vibration modeling a complex process. This study aims for application of an Artificial Neural Network (ANN) method and Geomechanical parameter relationships to simulate the blast-induced vibration for a Brazilian mining site and then compare them to the traditional approach. ANN had the best performance for this mine despite having demanded large datasets (as much as for the traditional approach), while geomechanical parameters like RQD and GSI may be used to deliver a fair approach even without seismic data. Also, ANN methods may be useful in dealing with a large amount of information to facilitate the simulation process when combined with other methods. Therefore, alternative prediction methods may be helpful for small budget mining operations in planning and controlling blast-induced vibration and helping mining in urban areas becoming a more sustainable activity.

Esse trabalho emprega o método para avaliar a abrasividade proposto por Golovanevskiy e Bearman (2008). Esse método, ensaio de abrasão por impacto deslizante (Gouging Abrasion Test), é realizado em condições de alta tensão/alto impacto de desgaste. O método consiste de uma ponteira cilíndrica com uma ponta cônica de 90º, que, em trajetória pendular, atinge uma amostra de rocha com energia de impacto de 300 J e velocidade da ordem de 5,2 m/s. O Gouging Abrasion Index (Gi) é calculado como sendo a média do diâmetro da ponta cônica, após desgaste, em milímetros e o resultado é multiplicado por 10. Esse trabalho verificou a adequabilidade do Gouging Abrasion Test, para um pequeno número de amostras de rocha, que representam, qualitativamente, os principais tipos de rocha encontrados em trabalhos de corte, perfuração e britagem no Brasil, e a sua correlação com outros ensaios consagrados como a resistência à compressão, o desgaste Amsler e a dureza Knoop. Essa análise mostrou alta correlação entre Gi e a dureza Knoop (R² = 0,94), baixa correlação com o desgaste Amsler (R² = 0,41) e nenhuma correlação com resistência à compressão uniaxial.

This paper employs the method for evaluating abrasion proposed by Golovanevskiy e Bearman (2008). This method, Gouging Abrasion Test, employs high-stress load gouging/sliding impact wear. The method consists of a steel wear tool with a 90º sharp conical tip. This tip attacks a rock sample in a swinging trajectory with an impact energy of 300 J and a speed of around 5.2 m/s. The Gouging Abrasion Index (Gi) is determined as 10 times the average diameter in millimeters of the conical tip (now flat) after one event of wear. This paper evaluates the Gouging Abrasion Test suitability in a small group of rocks that represents some of the main types to be found in Brazil´s rock cutting, drilling and crushing works. Its relationship to other usual tests like uniaxial compressive strength, Amsler abrasive wear and Knoop hardness were also verified. The results show high correlation between Gi and Knoop hardness (R² = 0,94), low correlation with Amlser wear (R² = 0,41) and no relation to uniaxial compressive strength.