摘要
针对火箭增压系统设计过程中关键换热参数难以确定的问题,提出1种基于混沌自适应粒子群算法(ACPSO)的增压系统换热参数辨识方法。建立了考虑换热模型的液体火箭增压系统非线性数学模型,通过灵敏度分析筛选出待识别参数,采用含有早熟抑制和惯性权重自适应调节策略的粒子群方法对选定参量进行辨识,对引入了权重衰减机制的方均根目标函数进行优化。结果表明:辨识后的仿真曲线与实测曲线具有良好的一致性,氧箱增压电磁阀打开时间仿真值与实测值偏差低于3%,气瓶1次飞行结束温度仿真值与实测值仅相差2.4 K。将辨识结果应用于某相似的新研型号,气瓶设计容积和质量相比绝热假设的设计方案降低了32 L和11.6 kg,有效减少了由设计过度冗余造成的额外试验和设计迭代成本。
In order to solve the difficulty of determining the key heat transfer parameters in the design of rocket pressurization system,a heat transfer parameter identification method based on adaptive chaotic particle swarm optimization(ACPSO)algorithm was proposed.The mathematical model of the pressurization system considering the heat transfer terms was established.The parameters to be identified were selected by sensitivity analysis,and then identified by particle swarm optimization method with local minimum prevention and adaptive weight strategy.The root mean square function with weight decay was optimized.The results showed that the identified simulation value was in good agreement with the measured value.The deviation between the simulation and the measured value of the opening time of the pressurization electric valve of the oxygen tank was less than 3%,and the deviation between the simulation value and the measured value of the temperature at the end of a flight of the gas bottle was only 2.4 K.If the identification results were applied to a similar newly developed pressurization system,the volume and weight of the gas bottle was reduced by 32 L and 11.6 kg compared with the design under the adiabatic assumption,which effectively saved cost caused by redundant design.
作者
张效溥
李志敏
徐鹏里
李春晓
王振剑
ZHANG Xiaopu;LI Zhimin;XU Pengli;LI Chunxiao;WANG Zhenjian(Shanghai Aerospace System Engineering Institute,China Aerospace Science and Technology Corporation,Shanghai 201109,China)
出处
《航空动力学报》
EI
CAS
CSCD
北大核心
2024年第4期224-233,共10页
Journal of Aerospace Power
关键词
系统工程
传热
粒子群算法
参数辨识
火箭增压系统
system engineering
heat transfer
particle swarm optimization algorithm
parameter identification
rocket pressurization system
