节点文献
基于机器学习的航空燃油换热器结焦特性研究
Machine Learning-based Study on Coking Characteristics in Fuel Heat Exchangers for Aero-engines
【摘要】 航空发动机用燃油–空气换热器在高温工况下面临结焦风险,导致其冷却性能下降。为探究燃油结焦对换热器热力性能的影响规律并支撑其性能修正设计,针对所提出的非均匀布置燃油–空气蛇形管换热器,采用三种机器学习算法构建了RP-3燃油结焦速率预测模型,建立了换热单元结焦厚度随燃油流动的计算模型,分析了不同工况下换热器性能与结焦特性的沿程变化规律,揭示了燃油结焦对流动与换热特性的影响机制,提出了减少结焦的措施。结果表明,运行10 h后,壁面结焦导致燃油侧热阻增大约3.7%;当质量流量由0.03 kg·s-1增大至0.05 kg·s-1时,结焦总量减少27.8%。
【Abstract】 Fuel-air heat exchangers in aero-engines face coking risks under high-temperature conditions, compromising their cooling performance. To investigate the influence of fuel coking on the thermal performance of heat exchangers and enable refined design, this study employs three machine learning algorithms to develop a coking rate prediction model for RP-3 aviation fuel. A computational model is established to characterize the evolution of coke deposit thickness along the channel in heat exchange units. The impact of fuel coking on flow and heat transfer characteristics is revealed, and the heat exchanger performance and coking behavior under various operating conditions is analyzed. Measures to reduce the coking rate are proposed. The results show that after 10 hours of operation, wall coking increases thermal resistance on the fuel side by approximately 3.7%. When mass flow rate increases from 0.03 kg·s-1 to 0.05 kg·s-1, the total coke deposition decreases by 27.8%.
【Key words】 machine learning; RP-3 fuel; coking rate; fuel-air heat exchanger; thermal resistance;
- 【文献出处】 工程热物理学报 ,Journal of Engineering Thermophysics , 编辑部邮箱 ,2026年02期
- 【分类号】V231;TP181
- 【下载频次】27