Genetic algorithm based optimization of nozzle profiles for a hydrogen turbo-expander

Liquid hydrogen plays an important role in the large scale storage and long-distance transportation. The development of hydrogen turbo-expander is the key to increase the efficiency of hydrogen liquefaction, reducing the cost of liquid hydrogen production, storage and transportation. In this paper, a numerical model of hydrogen nozzle is established and validated against experimental data. The performance of four traditional nozzle profiles in hydrogen turbo-expanders is simulated and analyzed. The poor uniformity of outlet flow angle is generally found in these nozzle profiles, leading to nozzle passage and impeller incidence losses in turbo-expanders. A genetic algorithm based method is proposed to optimize the nozzle profiles. The optimization objectives involve the nozzle efficiency, the uniformity of the nozzle outlet angle and the uniformity of the outlet Mach number. The deviations in the outlet angle and Mach number of the optimized nozzle are reduced by 50.26 % and 14.03 %, respectively, while the nozzle efficiency reaches 98.64 %. The matching characteristics of the optimized nozzle with the impeller are obtained via simulation of a hydrogen turbo-expander, and the results indicate the expansion efficiency can be increased by 1.53 %.