泄漏(经济)
液态氢
氢
环境科学
汽车工程
航空航天工程
计算机科学
航空发动机
石油工程
核工程
材料科学
工程类
机械工程
物理
宏观经济学
经济
量子力学
作者
Dimitrios Lamprakis,David John Rajendran,Mani Sekaran Santhanakrishnan,Seyfettin Coskun,Ioannis Roumeliotis,Vassilios Pachidis,M.K. Yates
标识
DOI:10.1115/gt2024-128734
摘要
Abstract A comprehensive operational characterisation of a representative, Liquid Hydrogen (LH2) aircraft engine pump, a key enabler for future hydrogen aviation, is presented in this work. The implications of leakage flows are investigated in a 2-stage, high-pressure pump for a wide range of flow rates and rotational speeds, through 3D (U)RANS simulations. The study compares two configurations: a baseline model comprising the primary flow path components — inducers, impellers and volutes and a realisable pump hardware that includes hub, shroud and power unit cavities. Performance metrics, including head changes and efficiencies, are extracted both at a component and system level. Leakage flow rates of 27.6% and up to 92.9% of the overall pump flow rate are recorded at design and lowest flow points respectively. The head loss in the mid to low flow rates does not exceed 4.5%, but the efficiency diminishes by up to 13.5% at off-design operation. The component analysis indicates significant penalties in impeller efficiency. At high flow rates, the presence of leakage flows improves the overall pump performance by 43% and 27% in head rise and efficiency, due to reduced losses in volutes and connecting ducts. The detailed characterisation of pump behaviour described in this work is of importance in development of safe, reliable and predictable design of aircraft LH2 pumps. These aircraft pumps are different from LH2 pumps utilized in rocketry and for cooling in nuclear industry due to the requirement to operate with wider turn-down ratios and often at low specific speeds. Therefore, this study addresses design considerations in this enabling technology that ensures the delivery of pre-conditioned fuel according to the aircraft operating conditions.
科研通智能强力驱动
Strongly Powered by AbleSci AI