Current Proceedings on Technology

Yazarlar: F. Nakhaei, M. Irannajad

Konular:-

Anahtar Kelimeler:Flotation column,Radial basis function,Recurrent neural network,Metallurgical performance

Özet: In recent years, artificial neural networks (ANNs) systems were employed for developing the predictive models to estimate the needed parameters. In this study, radial basis function neural network (RBFNN) and layer recurrent neural network (RNN) were applied to predict the performance of flotation column. Also, a review of NN applications for productivity studies was given together with a detailed discussion of the RBFNN and RNN models which were utilized in conducting the present research. In order to acquire the training data for the NNs, the case study was conducted at Sarcheshmeh copper complex pilot plant for the predictive performance evaluation of RBFNN and RNN models in estimating the Cu and Mo grades and recoveries in flotation column concentrate. Results show that, the constructed RNN exhibited a better performance than RBFNN for predicting Metallurgical performance of flotation column. The RNN model predicted the Cu and Mo grades with correlation coefficients of 0.91, 0.91, and their recoveries with correlation coefficients 0.88 and 0.87 respectively in testing stages. The modeling methodology proposed can be directly applied to develop tools for predicting the metallurgical performance in flotation column concentrate.

---