Abstract: Based on the features of melting process of regenerative aluminum melting furnaces, a mathematical model with user-developed burner reversing and burning capacity model and melting model, was established. Based on validating results by heat balance test for an aluminum melting furnace, CFD software FLUENT was used to simulate the coupling field between aluminum bath and combustion space. Considering influence analysis of burner arrangement on the performance of regenerative aluminum melting furnace, the relationship between burner arrangement and evaluation criteria was built using non-linear regression. Non-dominated sorting genetic algorithm-II was used to deal with multi-objective optimization for burner arrangement. The results show that the minimum RSD (relative standard deviation) of aluminum temperature (2.65%), dimensionless melting time (0.82) and RSD of furnace temperature (14.03%) could be obtained under the optimum conditions of vertical angle of burner 23.56o, height of burner 1471.81 mm, and horizontal angle between burners 62.05o.

Key words: regenerative aluminum melting furnace, melting process, numerical simulation, non-dominated sorting genetic algorithm-II, burner arrangement, multi-objective optimization, non-linear regression

摘要： 考虑蓄热式铝熔炼炉运行过程的特点，通过耦合用户自定义熔化模型和燃烧器换向及燃烧量变化模型，建立了蓄热式铝熔炼炉熔炼过程的数学模型. 运用计算流体力学软件FLUENT模拟了熔炼过程中炉膛和熔池耦合物理场，并对模拟结果进行了验证. 依据燃烧器布置方式对蓄热式铝熔炼炉熔炼性能的影响分析，以模拟结果为基础，建立了燃烧器布置方式与优化指标之间的非线性相关关系，再结合非支配排序遗传算法NSGA-II实现燃烧器布置方式的多目标优化. 优化结果表明，当燃烧器倾角23.56o、燃烧器高度1741.81 mm和燃烧器间水平夹角62.05o时，能获得较佳的铝料温度相对标准差(2.65%)、无量纲熔炼时间(0.82)和炉膛温度相对标准差(14.03%).

关键词: 蓄热式铝熔炼炉, 熔炼过程, 数值模拟, 支配排序遗传算法, 燃烧器布置方式, 多目标优化, 非线性回归