-
Kássia Fernanda da Silva
Programa de Pós-Graduação em Instrumentação, Controle e Automação de Processos de Mineração, Universidade Federal de Ouro Preto (UFOP) e Instituto Tecnológico Vale (ITV), Ouro Preto, MG
-
Thomás V. B. Pinto
Instituto Tecnológico Vale (ITV), Ouro Preto, MG
-
Thiago A. M. Euzébio
Helmholtz-Zentrum Dresden-Rossendorf
Keywords:
Mining, Artificial Intelligence, Machine Learning, Neural Network, SAG Mill, Energy Efficiency
Abstract
Predicting the energy consumption of a grinding mill in real-time is of great importance for a mineral processing plant. Grinding is a critical stage in mineral processing and the number one in electrical energy consumption. The mills represent more than 70% of the total energy consumed in some units. The consumption prediction enables a balance in the production demand in a way that prefers the more economically viable moments. Thus, it is an effort toward sustainable mineral processing and low-cost operation. This paper presents the initial study to model the energy consumption of a grinding mill with an artificial neural network. The industrial data are provided by a copper mine plant in the north of Brazil.
