基于ANN引导的NSGA-II优化在多联产太阳能系统中选择传热流体和相变材料

Abstract This research explores the selection of heat transfer fluids (HTFs) and phase change materials (PCMs) within a multi-generational energy system tailored to the climatic conditions of Bandar Abbas. The system incorporates parabolic trough collectors (PTCs), thermal energy storage (TES), and a high-temperature organic Rankine cycle (HTCORC). Three nanofluid-based HTFs (TiO 2 , Al 2 O 3 , CuO) are assessed for both PTCs and TES cycles, alongside two PCMs (NaOH–NaCl and MgCl 2 –KCl–NaCl) for TES. For the HTCORC unit, Toluene, Benzene, and Cyclopentene are considered as working fluids. To efficiently identify optimal HTFs and PCM, an artificial neural network-assisted NSGA-II algorithm is employed, targeting two objective functions: (1) total cost rate, exergy efficiency, and hydrogen production; and (2) total cost rate, power output, and energy efficiency. Three distinct Multi-criteria decision-making techniques (PROMETHEE, TOPSIS and VIKOR) are used to select the best trade-offs. For the first group of objective functions, the best configuration is Al 2 O 3 & MgCl 2 –KCl–NaCl & Benzene, yielding a cost rate of 36.0054 $/day, 12.72 % exergy efficiency, and 141,297.365 kg/year hydrogen production. For the second set, Al 2 O 3 & MgCl 2 –KCl–NaCl & Toluene performs best, with a cost rate of 37.2402 $/day, 247,386.12 W power output, and 24.62 % energy efficiency.