Numerical Simulation And Optimization Of Melting Process Of A Regenerative Aluminum Melting Furnace

Aluminum melting furnaces are key equipment in melting and casting factory, and a large amount of energy may be consumed during aluminum and aluminum alloy process procedure. In recent years, along with raised aluminum processing yield and demand of high quality for aluminum alloy castings in our country, high thermal efficiency, less pollutant emission and good product are demanded. Hence, it is important for both theories and practices to comprehensively understand melting process of aluminum melting furnaces by numerical simulation, which will result in great improvement of energy-saving and optimization in aluminum melting furnaces.According to the features of melting process of regenerative aluminum melting furnaces, in the present work, this paper concerns fuel combustion, heat transfer between combustion space and aluminum bath, phase change, oxide growth, burner reversing and varying of heating load, heat loss through furnace walls. A mathematical model with user-developed melting model, oxide loss model, burner reversing and burning capacity model was established. This paper presents numerical simulation of melting process of a regenerative aluminum melting furnace using hybrid programming method of FULENT UDF and FLUENT Scheme. Based on the principle of heat balance calculation and UML (Unified Modeling Language), heat balance calculation software for regenerative aluminum melting furnaces was developed. By employing heat balance test data in a company, heat balance calculation was conducted with the program and numerical results were further verified and effective measures for increasing heat efficiency of regenerative aluminum melting furnaces were put forward. The following conclusions can be derived:(1) The numerical results are basically in good agreement with the experiment results. The trend of simulation results is in accordance with that of measured data, they have similar distribution and relative error is less than5%. Thus, the computational models were proven to be reliable and accurate. The results show that melting phenomenon of the furnace may be revealed thoroughly. It is also indicated that optimization of parameters for aluminum melting furnaces may be studied by the above model.(2) Aluminum temperature increases slowly with melting time in solid-liquid zone, but rises faster when leaving solid-liquid phase lines. Furnace temperature firstly increases periodically with melting time, then decreases stepwise, lastly increases periodically. Oxide weight parabolically increases with melting time. As oxide porosity increases, the increase of aluminum temperature becomes slow.(3) RSD (relative standard deviation) of aluminum temperature firstly increases, then decreases, finally increases. RSD of furnace temperature firstly decreases periodically with melting time, then increases stepwise, finally decreases periodically. The heat flux through aluminum surface firstly increases with melting time, then decreases slowly.(4) In early melting stage, flue gas temperature decreases with liquid fraction, then increases in later melting stage. Oxygen concentration in flue gas increases with liquid fraction in early melting stage, then remains constant in later melting stage.With cause and effect diagram of performance for aluminum melting furnaces, vertical angle of burner (A/θ), height of burner (B/H), secondary flue (C), swirl number (D/S), horizontal angle between burners (E/α), air preheated temperature (F/T), natural gas mass flow (G/M), and air-fuel ratio (H/n) were selected to investigate their effects on the performance of aluminum melting furnaces. Based on the analysis of preliminary experimental tests, RSD of aluminum temperature (Y1), melting time (Y2) and RSD of furnace temperature (Y3) were selected as evaluation criteria, and this study used fuzzy judgment matrix and created a linear programming model in order to solve the weights for the evaluation criteria by MATLAB. CFD technique, in association with the Taguchi method was employed for parameter optimization of melting process of aluminum melting furnaces. The main conclusions are drawn as follows:(1) By non-linear regression on the base of simulation results, the empirical correlations for the RSD of aluminum temperature (Y1), melting time (Y2) and RSD of furnace temperature (Y3) are obtained as follows:(2) By statistical analysis of evaluation criteria such as RSD of aluminum temperature (Y1), melting time (Y2) and RSD of furnace temperature (Y3), the optimum condition of aluminum melting furnaces is A2B1C1D2E1F3G3H1. Through ANOVA (Analysis of Variance) and ANOM (Analysis of Means) for S/N ratio (signal-to-noise ratio) and mean value, the factors are classified into four categories as follows: important factors are D, E, F and G, robust factor is A, regulatory factor is H, and secondary factors are B and C. It is indicated that numerical simulation of aluminum melting furnaces is successful through the confirmation experiment. And the results are accurate and reproducible. The product quality and robustness of process development with the optimal parameter may be improved.(3) The rules of influence factors on performance of aluminum melting furnaces are achieved by numerical analysis.To reduce heat loss and save cost, three-layer slab was simplified for furnace linings of widely-used aluminum reverb furnaces in aluminum casting industry. Heat transfer analysis of different heat-insulating mode on furnace lining was carried out. Based on economic thickness method, furnace linings were optimized by computer programming. On this basis, a three dimensional mathematical model of aluminum melting furnaces including linings was developed. Furnace linings with40-week work system of before and after optimization were simulated by CFD software FLUENT. It is indicated that the results of optimization are successful and ideal economic effect is obtained.