Aluminum melting process is a semi-continuous process with high-energy consumption. In this paper, a software-based strategy which considers numerical simulation and control scheme simultaneously is employed to improve energy consumption of aluminum melting furnaces without changing the hardware. For numerical simulation, a nonlinear steady-state optimization is performed offline to obtain optimal operating points. Extensively, the optimal operating conditions include not only product exit temperature, but also ratio of combustion air flow and natural gas flow, furnace temperature, flue gas temperature and some important manipulated variables. For control scheme, a two-layer model predictive control which consists of steady state target calculation and dynamic optimization is developed to track the optimal operating conditions. In steady state target calculation layer, a priority strategy is proposed based on the different importance of controlled variables to make the steady state targets more reasonably. In dynamic optimization layer, a quadratic objective function is defined in terms of tracking both the optimal steady-state of controlled variables and manipulated variables. The method is successfully carried out in F aluminum melting furnace of a company in Tianjin. Compared with previous operation, the comprehensive energy consumption and the comprehensive energy consumption for unit output of product decrease 5.99% and 6.56% respectively.