Dimensionality reduction design and optimization of gaseous oxygen cooling radiator for in-situ lunar resources utilization

TY - JOUR

T1 - Dimensionality reduction design and optimization of gaseous oxygen cooling radiator for in-situ lunar resources utilization

AU - Yin, Pengyu

AU - Yao, Xiaoyu

AU - Zheng, Siyu

AU - He, Chengdan

AU - Wang, Yongjun

AU - Wang, Jin

AU - Shen, Jun

N1 - Publisher Copyright: © 2026 Elsevier Ltd.

PY - 2026/6/1

Y1 - 2026/6/1

N2 - The radiation cooling process of the oxygen on the Moon is crucial for the in-situ lunar resource utilization. Cooling power and mass are the most important performance indicators of the radiator, which are significantly affected by the radiator’s structure. It is crucial to design and optimize the cooling radiator structure. A dimensionality reduction modeling method (DRMM) and the corresponding mathematical model are proposed in this study, which can eliminate the effect of actual radiation area variation and reduce the computational load significantly during the optimization process. Based on this, the radiator structure optimization for cooling power and mass is conducted and validated by the combination of DRMM, Response Surface Methodology (RSM) and Multi-Objective Particle Swarm Optimization (MOPSO). The results show that among the structural parameters, the curve number has the most significant impact on both cooling power and mass, followed by vertical and horizontal edge distances. The final optimized structure of the radiator could reduce the oxygen temperature from 423.15 K to 299.46 K with almost no pressure drop. Moreover, it reduces the radiator mass by 13.57 % with an almost negligible decrease in cooling power compared to the original structure. The structure optimization based on the DRMM would be helpful for the radiator optimization in the context of in-situ lunar resource utilization.

AB - The radiation cooling process of the oxygen on the Moon is crucial for the in-situ lunar resource utilization. Cooling power and mass are the most important performance indicators of the radiator, which are significantly affected by the radiator’s structure. It is crucial to design and optimize the cooling radiator structure. A dimensionality reduction modeling method (DRMM) and the corresponding mathematical model are proposed in this study, which can eliminate the effect of actual radiation area variation and reduce the computational load significantly during the optimization process. Based on this, the radiator structure optimization for cooling power and mass is conducted and validated by the combination of DRMM, Response Surface Methodology (RSM) and Multi-Objective Particle Swarm Optimization (MOPSO). The results show that among the structural parameters, the curve number has the most significant impact on both cooling power and mass, followed by vertical and horizontal edge distances. The final optimized structure of the radiator could reduce the oxygen temperature from 423.15 K to 299.46 K with almost no pressure drop. Moreover, it reduces the radiator mass by 13.57 % with an almost negligible decrease in cooling power compared to the original structure. The structure optimization based on the DRMM would be helpful for the radiator optimization in the context of in-situ lunar resource utilization.

KW - Cooling radiator

KW - Dimensionality reduction

KW - In-situ

KW - Lunar resources

KW - Multi-objective optimization

KW - Oxygen

UR - https://www.scopus.com/pages/publications/105029675365

U2 - 10.1016/j.ijheatmasstransfer.2026.128473

DO - 10.1016/j.ijheatmasstransfer.2026.128473

M3 - Article

AN - SCOPUS:105029675365

SN - 0017-9310

VL - 260

JO - International Journal of Heat and Mass Transfer

JF - International Journal of Heat and Mass Transfer

M1 - 128473

ER -