基于响应面回归建模与遗传算法耦合的核能驱动集成制冷与发电循环的㶲经济分析与多目标优化

Abstract The current study explores the thermal and economic performance of an innovative combined cooling and power generation system integrating a reheat recompression main compression intercooling Supercritical CO 2 (sCO 2 ) cycle with a double effect absorption refrigeration cycle. To assess the effects of different input parameters on its performance, a detailed parametric study is conducted. The combined system has been modeled and proposed to harness 600 MW of thermal energy from the nuclear reactor. The dataset extracted from thermodynamic and exergoeconomic models has been utilized for response surface regression modeling (RSM) and its accuracy has been evaluated using different error matrices. Finally, multi-objective optimization has been conducted integrating the quadratic regression model with genetic algorithm (GA) on three objective functions: energy utilization factor (EUF), exergy efficiency (η ex ) and total product unit cost (c p,tot ) which provided 84 Pareto optimal datasets. Genetic algorithm and LINMAP are incorporated to select an ideal operating condition from the pareto optimal solutions. Single point optimization revealed that the novel cycle has a maximum EUF, and second law efficiency of 69.12 % and 77.07 % respectively with a minimum unit cost of 9.46 $/GJ. The cycle generates 400.4 MW of power and 116.2 MW of evaporative cooling when operated at basic design point. Key findings from this work demonstrate substantial performance enhancements in the integrated cycle compared to the conventional ones integrating the sCO 2 cycle with a single effect ARS. This research could significantly advance the harnessing of nuclear energy by optimizing advanced combined power and cooling cycles. Improving system efficiency and economic feasibility could pave the way for major advancements in nuclear power generation by introducing new areas for research and innovation.