Analysis and optimization of a nuclear power S-CO2 recompression Brayton CHP system on Mars using hybrid CPO-SVR-COA metaheuristics

Abstract

Among the planets Mars most closely mirrors Earth and is thus a more viable option for human habitation. Its exceptional strategic significance and value further highlight the crucial necessity of establishing a base on Mars. The primary challenge in constructing a Mars base lies in ensuring a stable supply of electricity and heat. Nuclear reactors, with their high power output, extended operational lifespan, and all-weather stability, serve as an optimal energy source for such a base. Moreover, the Brayton cycle within a megawatt-scale nuclear energy system can enhance efficiency in energy conversion. In this study, a supercritical carbon dioxide (S-CO₂) recompression Brayton cycle combined heat and power system (RCBC-CHP) was proposed in the context of Mars base, utilizing a nuclear reactor as its heat source. The thermodynamic model and mass estimation model were developed, analyzing the effects of key parameters—such as compressor inlet temperature and pressure, turbine inlet temperature, pressure ratio, and splitting ratio—on total thermal efficiency, power generation efficiency, total mass, and heating to power ratio. The system prediction model was developed using support vector regression (SVR) optimized by the crested porcupine optimizer (CPO), forming the CPO-SVR approach. Simultaneously, the coati optimization algorithm (COA) was employed to optimize the system, with total thermal efficiency, power generation efficiency, and total mass set as the optimization objectives. Balancing the goals of maximizing total thermal efficiency and power generation efficiency while minimizing total mass, the optimal results obtained were a total thermal efficiency of 79.40 %, a power generation efficiency of 38.16 %, a total mass of 19631.37 kg, and a heating to power ratio of 1.0434.