A fresh view on particle swarm optimization to develop a precise model for predicting rock fragmentation

Abstract

Purpose The purpose of this paper is to propose a gauge for the convergence of the deterministic particle swarm optimization (PSO) algorithm to obtain an optimum upper bound for PSO algorithm and also developing a precise equation for predicting the rock fragmentation, as important aims in surface mines. Design/methodology/approach In this study, a database including 80 sets of data was collected from 80 blasting events in Shur river dam region, in Iran. The values of maximum charge per delay (W), burden (B), spacing (S), stemming (ST), powder factor (PF), rock mass rating (RMR) and D80, as a standard for evaluating the fragmentation, were measured. To check the performance of the proposed PSO models, artificial neural network was also developed. Accuracy of the developed models was evaluated using several statistical evaluation criteria, such as variance account for, R-square (R2) and root mean square error. Findings Finding the upper bounds for the difference between the position and the best position of particles in PSO algorithm and also developing a precise equation for predicting the rock fragmentation, as important aims in surface mines. Originality/value For the first time, the convergence of the deterministic PSO is studied in this study without using the stagnation or the weak chaotic assumption. The authors also studied application of PSO inpredicting rock fragmentation.