Optimization of a closed Brayton cycle for space power systems

Abstract

This paper describes an optimization study of the space power system based on a closed Brayton cycle (CBC). The objective of the work is to minimize the total mass of power system through optimizing the critical parameters of system components. Non-dominated Sorting Genetic Algorithm (NSGA-II) was used to optimize the total system mass, together with performance calculation of the CBC system and the design modules of the components. Component mass predicted by the regenerator module developed in this work shows a good agreement with empirical data. The optimal values of the critical parameters in the Pareto Front, corresponding to designs with smallest system mass at specific turbine inlet temperatures, were presented and discussed. Furthermore, a sensitivity analysis of the parameters was conducted using Garson Algorithm. Results show that an increase of 4% of the turbine inlet temperature from 1150 K results in a decrease of 6% of the system mass. It is also suggested that the regenerator heat effectiveness is a key design parameter to minimize the system mass. The analysis presents the importance of detailed design variables of components in a CBC system design.