Towards improved energy efficiency in the electrical connections of Hall–Héroult cells through Finite Element Analysis (FEA) modeling

Abstract

The Hall–Héroult smelting process is the most widely used method for extracting aluminum from alumina through electrolytic reduction. The large quantity of electrical current employed in the process and the typically large-scale output quantities combine to make aluminum smelting a high energy consuming industry. Therefore initiatives aimed at reducing energy losses in this sector are currently of great environmental, commercial and political interest. Stricter emissions regulations and escalating electricity costs continue to provide the driving force towards making primary production of aluminum cleaner and more sustainable in the years to come. Multiphysics modeling aimed at optimizing the process can play an enabling role in advancing energy-efficient solutions for this industry. In this article the authors describe two advanced finite element models relating to the anode assembly that may be used for designing energy efficient electrical connections. The value of using such in-depth and realistic models is demonstrated through a discussion of the significance of the various predictions in relation to energy efficiency. The relative influences of various parameters on total voltage drop across the anode assembly are quantified and the corresponding effects on energy consumption and greenhouse gas emissions are calculated. Areas for redesign aimed at improving energy efficiency are identified. These models may be adapted for related energy-intensive processes, e.g., smelting of other metals or in metal-winning. A small number of smelters owned by leading multinationals are already utilizing modeling to improve their processes. It is argued that the wider smelting industry should consider making Finite Element Analysis (FEA) modeling an integral part of their design phase as well as trouble-shooting initiatives so as to be able to make informed decisions that are based on quantitative scientific data.