Thermoeconomic multiobjective optimization of tobacco drying heat pump recovering waste heat from monocrystal silicon furnace based on SVR ANN model in Southwest China

Abstract

Abstract Great quantities of energy are consumed by heat pump drying systems (HPDSs), such as tobacco drying. It can be reduced by increasing evaporating temperature using waste heat. In this paper, based on the case of two main industries in southwest China (monocrystalline silicon industry and tobacco industry), a novel integrated HPDS especially for the tobacco drying using waste heat from single crystal silicon furnaces is proposed. It is then analyzed from an energetic and economical perspective, investigating the effects of heat exchanger areas on system performance. To simultaneously minimize system CO2 emissions and total annual cost (TAC), a multiobjective optimization using nondominated sorting genetic algorithm is conducted based on the support vector regression model, which has been rarely reported for integrated HPDS improved by waste heat. The final optimum solution on the Pareto optimal front is selected using different decision‐making methods (LINMAP, TOPSIS, and fuzzy decision method), and are compared with the performance of the initial system. The results show that the TOPSIS method is most suitable in three decision‐making methods with a great reduction in m C O 2 ${m}_{C{O}_{2}}$ (−12.7%) and nearly unchanged in TAC (+0.5%) compared with the initial solution. And the decreases in m C O 2 ${m}_{C{O}_{2}}$ and TAC of the optimized system is as large as 50% and 1/3, compared with traditional vapor compression heat pump dryer without heat recovery of the monocrystal silicon furnace.

Keywords

• ANN
• energy saving
• heat pump dryer
• single‐crystal silicon
• waste heat