关键词: 塔式光热电站, 布雷顿循环, 基于S-CO₂的混合工质, 智能进化算法, 热力学性能, 参数优化, 典型日性能

Abstract:

To collaboratively optimize the structure and parameters of the Brayton cycle system in tower-type concentrating solar power plants， the effects of cycle configuration， operating parameters， and CO₂-based binary working fluids on the thermodynamic performance of a solar power tower coupled with a supercritical CO₂ Brayton cycle were investigated， aiming to identify suitable configurations and working fluids under different ambient-temperature conditions. A bi-level optimization framework combining a genetic algorithm with sequential quadratic programming was developed. The genetic algorithm searched cycle configurations and discrete variables， whereas sequential quadratic programming optimized continuous operating parameters，such as temperature and pressure. After validation against a published case， the method was applied to cycles with optional reheating， intercooling， and split-flow processes. Based on REFPROP 9.1， four groups comprising 40 CO₂-based binary mixtures were constructed using H₂S， He， butane， and krypton as additives. Their specific work and thermal efficiency were compared， and S-CO₂，S-CO₂-50% He， and S-CO₂-50% H₂S were selected for typical-day analyses under the meteorological conditions of Delingha. The proposed bi-level optimization effectively coordinated cycle configuration and parameter optimization. Compared with the reference case， the optimized cycle increased specific work by 24.6% and thermal efficiency by 2.12%， and the optimal configuration included reheating and intercooling without splitting flow. Comparisons of pure working fluids showed that， relative to S-CO₂， He and H₂S increased specific work by approximately 324.8% and 28.5%， respectively， whereas butane and krypton reduced it by 35.9% and 71.5%. The duration of high-specific-work operation was strongly affected by meteorological conditions and generally increased from the spring equinox to the winter solstice. During high-temperature periods， the specific work of S-CO₂-50%He reaches 286.1 kJ/kg， which was significantly higher than that of S-CO₂-50%H₂S （138.2 kJ/kg） and pure S-CO₂（119.4 kJ/kg）. However， under low-temperature conditions， the specific work of S-CO₂-50% He fell below that of S-CO₂， whereas S-CO₂-50% H₂S provided gains throughout the entire operating period. The genetic algorithm-sequential quadratic programming framework is effective for coupled optimization of cycle structure and operating parameters. Reheating and intercooling are beneficial for enhancing cycle performance， while flow splitting is unnecessary in the optimal design. S-CO₂-50% He is more suitable for high-temperature conditions requiring maximum specific work， whereas S-CO₂-50% H₂S offers more stable performance over a wider temperature range. Working-fluid selection should therefore comprehensively consider local meteorological conditions， off-design performance， and the duration of high-specific-work operation， rather than relying solely on design-point performance.

Key words: tower-type solar thermal power plant, Brayton cycle, S-CO₂-based mixed working fluid, intelligent evolutionary algorithm, thermodynamic performance, parametric optimization, typical day performance