Techno-economic and advanced exergy analysis and machine-learning-based multi-objective optimization of the combined supercritical CO2 and organic flash cycles

• Published in: Applied thermal engineering 2025-01, Vol.258, p.124667, Article 124667
• Main Authors: Sardroud, Ramin Ghiami, Mahmoudi, Seyed Mohammad Seyed, Ghasemzadeh, Nima, Avazpour, Mahyar, Nami, Hossein
• Format: Article
• Language: English
• Subjects: Advanced exergy; Exergoenvironmental; Machine learning optimization; OFC; Supercritical CO2 Brayton cycle; Techno-economic
• ISSN: 1359-4311
• DOI: 10.1016/j.applthermaleng.2024.124667

•The organic flash cycle is used as the bottom cycle for S-CO2 cycle.•Thermodynamic, techno-economic, and advanced exergy analyses are applied.•Machine learning-based multi-objective optimization is applied.•The exergy efficiency of 70.8% is reachable under the optimized conditions.•Under the optimized conditions, the net present value is enhanced by 68.9% In the past few years, attention has been directed towards the advancement of renewable-driven supercritical CO2 recompression Brayton (SCRB) cycles. This highlights the impacts of sustainable and efficient power production towards transition in the electricity sector. The combination of the SCRB cycle with the organic flash cycle (OFC) is comprehensively investigated from thermodynamic, techno-economic, environmental, and advanced exergy perspectives. Conventional thermodynamic analysis alone cannot provide information on the cause of irreversibilities, the interconnections between components, and the possibility of enhancing their performance; therefore, advanced exergy analysis is implemented. Furthermore, an examination has been conducted into the effects of different decision parameters on the exergoenvironmental indicators. Multi-objective optimization based on machine learning is performed on the SCRB/OFC plant to maximize exergy efficiency, net present value, and net power output. Results from advanced exergy analysis revealed that the reactor has the worst performance, with just 2.8 % of its total exergy destruction being avoidable. Furthermore, it is feasible to decrease its total exergy destruction rate by 21.6 % through enhancements in the performances of other components. Under the optimized conditions, exergy efficiency, net present value, and net power output are found to be 70.8 %, 1347.1 M$, and 306.4 MW, respectively, representing a 21.1 %, 68.9 %, and 21.1 % enhancement compared with the base conditions, respectively.