Influence of Meteorological Parameters on Explosive Charge and Stemming Length Predictions in Clay Soil during Blasting Using Artificial Neural Networks

Karlo Leskovar,Ranko Biondić,Denis Težak,Josip Mesec

doi:10.3390/app11167317

Karlo Leskovar, Ranko Biondić + Show 2 more

Open Access

https://doi.org/10.3390/app11167317

Copy DOI

Journal: Applied sciences	Publication Date: Aug 9, 2021
Citations: 2	License type: CC BY 4.0

Affiliation: University of Zagreb

Abstract

The influence of the meteorological parameters (precipitation and air temperature) during blasting in clay has a direct impact on the success of blasting. In the case of large amounts of precipitation (rain and snow) recorded in the subject area, blasting in clays cannot be carried out due to the grain of the clay and the inability to access the subject area. Moreover, the air temperature in the subject area affects the blasting performance. The most ideal temperature for blasting in clays is between 15 and 25 °C because then the clay has the best geotechnical characteristics. The research was conducted on the exploitation field Cukavec II, which is located near the city of Varaždin in the Republic of Croatia. Amount of precipitation and air temperature were considered to obtain the best blasting effect. Influence of meteorological parameters on the amount of the explosive charge and stemming length when blasting in clays was demonstrated via models based on Artificial Neural Networks (ANN). The ANN model network consists of a Long Short-term Memory (LSTM) part to process time dependent meteorological data, and fully connected layers to process blasting input data. Two types of explosive charges were compared, Pakaex and Permonex V19.

Highlights

Research conducted in the exploitation field Cukavec II (Figure 1) in the period from 2014 to 2016, proved the dependence of the amount of explosive charge and stem length on the formation of the volume of the spherical expansion, resulting from the detonation of the explosive charge in a borehole of 131 mm in diameter
The data is divided on three time periods, where the blasting results from 12 June 2014 and 20 July 2016 served as training data for the models, 12 June 2015 presents the validation dataset, and 31 August 2016 the testing dataset, respectively
The charts of explosive charge mass versus the resulting borehole expansion (Figures 6 and 7) and volume of resulting cavity (Figures 8 and 9) clearly indicate that it is possible to use Artificial Neural Networks (ANN)-s to predict the required mass of explosive charge, considering a desired expansion of the borehole or volume of the resulting cavity

Summary

Introduction

Research conducted in the exploitation field Cukavec II (Figure 1) in the period from 2014 to 2016, proved the dependence of the amount of explosive charge and stem length on the formation of the volume of the spherical expansion, resulting from the detonation of the explosive charge in a borehole of 131 mm in diameter. The main goal was to expand the knowledge about the possibilities of using explosive charges in geotechnical practice. This especially refers to blasting in clay soil by which spheres or cavities of other shapes are formed at different depths below the soil surface by activating a certain type and mass of explosives [1,2]. Previous research was conducted with the goal of examining geotechnical properties of the clay soil, where cohesion c, internal friction angle φ and volume weight γ were determined. Due to the characteristics of hydroalumosilicate clays, which means that it absorbs water and becomes brittle, blasting is directly affected [2,3]

Methods

Results

Conclusion