摘要
为兼顾固体推进剂的高比冲与经济性,提出一种多目标配方优化设计方法,并开发了相应的辅助设计程序。采用动态径向基(RBF)神经网络构建代理模型,以表征推进剂组分配比与比冲性能之间的非线性映射关系;结合非支配排序遗传算法(NSGA-Ⅱ)进行全局优化搜索,获取比冲与成本之间的帕累托(Pareto)最优解集。基于标准测试函数的评估结果表明,所提方法在相同模型调用次数下,解的分布性与收敛性较NSGA-Ⅱ提升86.49%;同时,在保持全局收敛性的条件下,高保真模型调用次数减少71.43%。在典型丁羧推进剂配方中的应用结果显示,成本最优解在降低成本10.18%的同时,理论比冲提升15.52%;性能最优解在比冲提升17.81%的条件下,成本亦降低9.68%,验证了该方法在推进剂优化设计中的有效性与实用性。
To balance the trade-off between high specific impulse and cost-effectiveness in solid propellants,a multi-objective formulation optimization design method was proposed,accompanied by the development of an auxiliary design program.The method employed a dynamic radial basis function(RBF)neural network to construct a surrogate model characterizing the nonlinear mapping between propellant component ratios and specific impulse performance.Combined with the non-dominated sorting genetic algorithm(NSGA-Ⅱ),it performed global optimization to obtain the Pareto optimal set between specific impulse and cost.Evaluation based on standard test functions indicated that,under equal model evaluation counts,this method improved solution diversity and convergence by 86.49%compared to NSGA-Ⅱ;moreover,it reduced the number of high-fidelity model evaluations by 71.43%while maintaining global convergence.Application to a typical carboxyl-terminated polybutadiene propellant formulation demonstrated that the cost-optimal solution achieved a 10.18%cost reduction accompanied by a 15.52%increase in theoretical specific impulse,while the performance-optimal solution attained a 17.81%increase in specific impulse with a 9.68%cost reduction.These results verify the method’s effectiveness and practical applicability in propellant optimization design.
作者
范馨平
魏然
任加忍
孙林
胡少青
徐司雨
FAN Xinping;WEI Ran;REN Jiaren;SUN Lin;HU Shaoqing;XU Siyu(National Key Laboratory of Solid Rocket Propulsion,Northwestern Polytechnical University,Xi'an 710072,China;Xi'an Modern Chemistry Research Institute,Xi'an 710065,China)
出处
《固体火箭技术》
北大核心
2025年第6期853-860,共8页
Journal of Solid Rocket Technology
基金
国家自然科学基金青年科学基金项目(52302479)。
关键词
成本效益
比冲
推进剂配方
优化
代理模型
cost-effectiveness
specific impulse
propellant formulation
optimization
proxy model
