INTELLIGENT CONTROL OF ALUMINIUM REDUCTION CELLS USING PREDICTIVE AND PATTERN RECOGNITION TECHNIQUES

摘要

A neural control logic scheme based on prediction and pattern recognition techni ques is used to control electrochemical processes such as aluminum electrolytic cells. T he predictive capacity of feedforward neural networks is used to predict the future values of decision variables to be used by the cell's control logic, enabling the control logic to apply anti cipated actions to cells at different states, thus avoiding anode effects and improving cell stabil ity. The patternrecognition capacity of LVQ-type neural networks is used to provide a closed-loo p control structure to the feeding of the cell as a function of cell resistance, alumina c oncentration and cell state. The closed-loop control structure enables the cell to operate at a near-o ptimal regime regardless of cell state.