Dynamic performance and multi-objective optimization of fin-concrete heat sinks

In a passive nuclear plant, the main control room must provide self-support functions for 72 h after an accident occurs. In this isolation mode, the heat generated by indoor occupants and equipment should be removed by using concrete walls that are 610 mm thick, and the finned ceiling made of concrete should maintain thermal habitability. However, the finned ceiling design has not been studied in normal buildings, and the cooling storage effect of plate-shaped fins is usually ignored during the design phase. In this study, a novel parametric resistance-capacitance thermal network is proposed to simulate the thermal performance of different fin-plate layout forms in the case of an accident. An experiment was conducted on a 1200 mm x 1200 mm x 610 mm finconcrete module for 72 h to validate the model under constant air temperature conditions. Based on the model, multi-objective optimization was conducted to obtain optimal solution sets by using a radial basis function approximation model. The results show that the optimal performance indicators can be classified into unsteady and quasi-steady states, in which the dominant factors for heat transfer include the heat transfer area and the thickness of the finned plate, respectively. Further, the optimal solution sets for different convection and heating conditions obtained through an optimization process are used to design a fin layout based on actual temperature control demands and cooling storage.