超声速飞行器综合热管理系统优化设计

Optimal Design of Comprehensive Thermal Management System for Supersonic Vehicle

针对飞行器超声速巡航时遭受机体外部气动加热与辐射换热的情况,提出了一套综合考虑燃油系统、冲压空气与消耗性冷却剂制冷系统及热防护系统的超声速飞行器综合热管理系统;基于二阶多项式响应面代理模型技术建立了综合热管理系统优化设计流程,并针对典型的超声速飞行包线,以热防护层厚度、回油质量流量以及消耗性相变冷却剂质量流量为设计变量,以燃烧室进口燃油温度为约束条件,以变量引起的起飞重量增量及其燃油质量代偿损失最小为设计目标,采用自适应模拟退火算法对热管理系统进行了优化设计。结果表明:采用二阶多项式响应面代理模型计算的结果与仿真计算结果的误差低于2%;在本文选取的设计变量取值范围内,最优方案倾向于热防护层厚度和冷却剂质量流量取最小值而回油质量流量取中间值,且最优方案较第一组初始方案就目标函数而言减重约16%。

In view of the fact that the aerodynamic heat and radiation heat transfer outside the body are used as the external heat sources in the supersonic cruise,an integrated thermal management system including fuel system,ram air,consumable phase change coolant cooling system and thermal protection system was proposed for the supersonic vehicle considering internal and external heat sources.An optimization design scheme of the comprehensive thermal management system was also proposed based on the technology of the quadratic response surface surrogate model.Then,the comprehensive thermal management system was optimized by using the adaptive simulated annealing algorithms,while the thickness of the thermal protective layer,the mass flow rate of the recycle fuel and the mass flow rate of the consumable phase change coolant were treated as design variables.The temperature of the fuel at the inlet of the combustor was treated as constraint condition;and the minimum increment of the takeoff weight and its fuel compensation loss caused by the variables were treated as objective function.Results showed that the error between the objective calculated by using the quadratic response surface surrogate model and commercial simulation software was less than 2%.In the range of design variables selected in this paper,the optimal scheme tended to the minimum value of the thickness of the thermal protective layer and the mass flow rate of the consumable coolant,but a middle value of the mass flow rate of the recycle fuel.In addition,comparing with the first scheme of the initial schemes,the optimal scheme reduced weight by about 16%in terms of the objective function.