The study of heat-mass transfer characteristics and multi-objective optimization on electric arc furnace

It is crucial to reveal the energy conversion and heat and mass transfer characteristics in the electric arc furnaces (EAFs). In this paper, a transient mathematical model was built to adequately capture the distribution of multi-physical fields in EAF. The numerical modeling method was verified against brown corundum smelting experiment data. In addition, the influence on the brawn corundum smelting process of the typical operating parameters has been examined, and the major problems which limit the performance improvement of smelting were analyzed. Moreover, four indicators were proposed to comprehensively evaluate the EAF performance, consisting of liquid volume fraction β, energy efficiency η, temperature rise index ς, and temperature uniformity index ξ. Furthermore, the optimal combination of operating parameters was selected automatically based on the back propagation (BP) neural network predicting model and non-domestic sorting genetic algorithm II (NSGA-II). The results showed that the local overheating phenomenon is serious and that homogenizing the energy distribution in the furnace is the key to improving the melting performance. Increasing electrode immersion and voltage can significantly increase the rate of heat and mass transfer. In the case of ψ = 0.75, υ = 0.6, when χ is increased from 0.2 to 0.6, τ_max is shortened by 13.3% and η_max is increased by 12.0%. While with ψ = 0.75 and χ = 0.4, τ_max is reduced by 36.0% when υ increases from 0.2 to 1.0, but η_max is also decreased by 7.0%. It was found that decreasing the height diameter ratio will improve the overall EAF smelting performance, but it also increases the area of heat loss. In the case of χ = 0.4, υ = 0.6, ξ_ave and η_max improved by 3.1 and 14.5%, respectively, when ψ decreases from 1.0 to 0.5. The EAF performances are considerably enhanced in comparison to the operating basic parameters (ψ = 0.75, χ = 0.4 and υ = 0.6), subject to optimal operation parameters (ψ = 0.61, χ = 0.59 and υ = 0.59). In this case, τ_max and ς_max reduced by 6.2 and 4.1%, respectively, while η_max and ξ_ave increased by 6.8 and 1.0%, respectively.