Predictive control of aluminum electrolytic cells using neural networks

Abstract

In this work, neural networks are built and trained to be used in a predictive control scheme for the aluminum electrolytic cell. An efficient control of the cell requires the knowledge of predicted future values of the decision variables in order to enable the standard (nonpredictive) control logic to take anticipated actions to prevent the anode effect, a destabilizing event occurring during cell operation. The networks are first trained on data obtained from a computer simulator of the cell prior to undergoing further on-line learning. Trained to predict the cell resistance and the resistance's trend indicators, the networks are applied to the control of cells in different cell states, with a view to preventing anode effects, the latter being deliberately induced by reducing the alumina feed rate or reducing the feeding frequency and duration. Results show that, with neural-predictive control, anode effects can be avoided, which should result in increased thermal stability, decreased power consumption, and reduced fluoride emissions. Further, the on-line learning capacity of the networks offers a good perspective for their application to other complex industrial processes as well.

A Monitoring System for the Copper Solvent Extraction - Electrowinning Process

“…The group has continued their research, and a fault diagnosis system for the zinc leaching process has also been published. (Wu et al, 2002) Application of model predictive control to uranium solvent extraction has been reported by Harper (1995), and to aluminium electrowinning by Meghlaoui et al (1998). Increased production levels, improved product quality and a reduction in the production costs as a result of advanced process control were reported in every case.…”

Development of Techniques and Tools for the Determination of Carbon Anode Quality

“…Berezin et al [1] used ANN to predict problems in an aluminum reduction cell. Meghlaoui et al [19] developed an ANN model for efficient control of alumina feeding in the electrolytic cell. Bhattacharyay et al [18,20,21] have studied the effect of different parameters on anode properties.…”

Effect of Coke Granulometry on the Properties of Carbon Anodes based on Experimental Study and ANN Analysis

“…Artificial neural network is used for predicting the values of dependent parameters for which no mathematical relation is available (Parthiban, 2007) or even though some mathematical relationship is available, it is hard to find the numerical parameters (Milewski, 2009). ANN models are widely used in various research fields including quality control (Bahlmann, 1999;Pang, 2004;Fruhwirth, 2007;Parthiban, 2007;Saengrung, 2007;Shang, 2008;Piuleac, 2010;Bhagavatula, 2012), prediction of compositions and properties of metallic and nonmetallic compounds (Wang, 2008;Asadi-Eydivand, 2014;Mohanty, 2014), aluminum reduction cell (Meghlaoui, 1998;Biedler, 2002;Boadu, 2010). However, there are only a few studies available (Berezin, 2002;Bhattacharyay, 2013;Bhattacharyay, 2015) in literature, which are directly related to carbon anodes used for the production of primary aluminum.…”