Name
Abstract | ||
|---|---|---|
One of the most important aspects of metal mine design is to determine the optimum cut-off grades and milling grades which relate to the economic efficiency of enterprises and the service life of mines. This paper proposes a hybrid intelligent framework which is based on stochastic simulations and regression, artificial neural network, and genetic algorithms is employed for grade optimization. Firstly, stochastic simulation and regression are used to simulate the uncertainty relations between cut-off grade and the loss rate. Secondly, BP and RBF network are applied to establish two complex relationships from the four variables of cut-off grade, milling grade, geological grade, and recoverable reserves to lost rate and total cost, respectively, in which, BP is used for the one of lost rate, and RBF is for the other. Meanwhile, the real-coding genetic algorithm is performed to search the optimal grades (cut-off grade and milling grade) and the weights of neural networks globally. Finally, the model has been applied to optimize grades of Daye Iron Mine. The results show there are 6. 6978 milling Yuan added compare to unoptimized grades. |
Year | DOI | Venue |
|---|---|---|
2012 | 10.1007/s00521-011-0593-7 | Neural Computing and Applications |
Keywords | DocType | Volume |
grade optimization,optimal grade,hybrid intelligent optimization method,optimum cut-off grade,loss rate,cut-off grade,multiple metal grades optimization,multiple metal gradescut-off grade � hybrid intelligentartificial neural networks � genetic algorithmsoptimization,stochastic simulation,lost rate,rbf network,unoptimized grade,geological grade,genetic algorithms,optimization,artificial neural networks | Journal | 21 |
Issue | ISSN | Citations |
6 | 1433-3058 | 1 |
PageRank | References | Authors |
0.38 | 12 | 3 |