Title: Simultaneous prediction of blast-induced flyrock and fragmentation in opencast limestone mines using back propagation neural network

Authors: Ratnesh Trivedi; T.N. Singh; A.K. Raina

Addresses: CSIR-Central Institute of Mining and Fuel Research, Regional Centre, Nagpur, 440006, India ' Department of Earth Sciences, Indian Institute of Technology Bombay, Powai, Mumbai, 400 076, India ' CSIR-Central Institute of Mining and Fuel Research, Regional Centre, Nagpur, 440006, India

Abstract: Goal of blasting operations is to achieve desired fragment size to operate the mine and plant economically while maintaining safety that includes prevention of flyrock accidents. This paper focuses on the simultaneous prediction of flyrock distance and fragmentation using back propagation neural network techniques. Thus, linear charge concentration, burden, spacing, stemming length, specific charge, unconfined compressive strength and rock quality designation are taken as input. Flyrock distance and fragment size are chosen as output. The predicted outputs by back propagation neural network (BPNN), multi variate regression analysis (MVRA) have been compared. The quite lower root mean square error (RMSE) and mean absolute error (MAE) in BPNN than MVRA prove that BPNN is a better prediction method. Also, the predicted output in BPNN correlates better with the observed output than MVRA. Sensitivity analysis for both independent variables for BPNN and MVRA is also included in this paper.

Keywords: opencast mining; blasting; BPNNs; back propagation neural networks; multiple regression; flyrock fragmentation; limestone mines; linear charge concentration; burden; spacing; stemming length; specific charge; compressive strength; rock quality; flyrock distance; fragment size; root mean square error; RMSE; mean absolute error; MAE.

DOI: 10.1504/IJMME.2016.078350

International Journal of Mining and Mineral Engineering, 2016 Vol.7 No.3, pp.237 - 252

Accepted: 31 Jan 2016
Published online: 15 Aug 2016 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article