A multi-factor comprehensive optimization of a supercritical carbon dioxide radial inflow turbine with low specific speed

An integrated method combining one-dimensional design and automatic three-dimensional optimization is proposed for SCO₂ radial inflow turbines with low specific speed. The optimization of the nozzle and impeller are performed simultaneously. Four optimization algorithms, namely, grey wolf optimizer, elephant herding optimization, genetic algorithm, and simulated annealing algorithm, are integrated with computational fluid dynamics simulation and finite element analysis to improve the turbine efficiency. Two constraints (mass flow variation and maximum stress) are imposed in the optimization process. The results indicate that the grey wolf optimizer is the optimal algorithm. The total-to-static efficiency of the optimal turbine is 89.56 %, which is increased by 3.25 %. Moreover, splitter blades are also investigated and optimized. The maximum total-to-static efficiency of 90.16 % can be obtained by reasonably arranging splitter blades. The proposed method is versatile, nimble, and easy to implement. It can improve the design efficiency and provide a geometric reference for low specific speed SCO₂ turbines.